ISSN:0003-2654    

DOI:10.1039/AN95378FX043

出版商:RSC    

年代:1953

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95378BX045

出版商:RSC    

年代:1953

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95378BP111

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

SEPTEMBER, 1953 Vol. 78, No. 930 The Determination of Ergosterol in Yeast Part I. The Ultra-violet Absorption of Purified Ergosterol BY W. H. C. SHAW AND J. P. JEFFERIES (Presented at the meeting of the Society on Wednesday, May 20th, 1953) Most published records agree on the wavelengths of the three main peaks in the ultra-violet absorption curve of ergosterol, but there are dis- crepancies in the extinction values reported for the various maxima, apparently because of the difficulty of preparing pure ergosterol by direct recrystallisation of the commercial material and to a lesser extent because of small variations in the moisture content of tlie hydrated sterol. Simple recrystallisation does not always yield a pure product; the best means of purification is by re- crystallising a suitable ester, with subsequent regeneration of the sterol.The preparation and purification of ergosterol benzoate is described, and the physical properties and ultra-violet absorption (in various solvents) of a purified specimen of ergosterol are recorded. METHODS for determining small amounts of ergosterol are mostly based either on colour reactions or on measurements of ultra-violet absorption. A number of colour reactions are k n ~ w n ~ s ~ , ~ and some can be applied quantitatively to pure solutions. Similar colours, however, are given by many other sterols. The colorimetric method with acetic anhydride and zinc ~ h l o r i d e , ~ , ~ modified by Pesez and Herb,zin,6 appears to be specific for ergosterol amongst the yeast sterols, but this method was found to give insufficiently reproducible results, even when applied to purified ergosterol.The most satisfactory method of detecting and determining small amounts of ergosterol is based on ultra-violet absorption measurements. The absorption of ergosterol is of high intensity and characteristic of A-5 :7 unsaturated sterols, amongst which 7-dehydrocholesterol7 is the only other one of importance known to occur naturally. Well-defined maxima occur a t 271.5, 282 and 293-5 mp in absolute alcohol, with a marked inflection at about 263 mp 509510 SHAW AND JEFFEKIEij : THE TlETERMINATION OF [Vol. 78 and a smaller inflection in the region of 263 mp (Fig. 1). Most literature reports agree on the position of the maxima, but discrepant results are found for the extinction value oi the pure compound at the main maximum (Table I).The methods of purifying the samples used in obtaining the figures in Table I require some comment. No details of methods are given in the first three publications mentioned. Wavelength, mp Fig. 1. Absorption curves. Curve A, ergosterol in absolute Con- alcohol; curve B, ergosterol benzoate in chloroform. centrations approximately 0.002 per cent. w/v Hogness et aZ.ll recrystallised ergosterol from ethyl alcohol and benzene and then twice from isooctane. Subsequently, Huber et aL7 showed that Hogness's values were too low and attributed this to decomposition during the several recrystallisations. They stated that the extinction values were highest on rccrystallising the commercial material once only from a mixture of alcohol and benzene.Lamb, Mueller and Beach,12 unable to confirm this, preferred rccrystallisation from acetone containing 1 per cent. of water ancl obtained results about 4 per cent. higher. It thus appeared desirable to investigate again the extinction values in the solvents it was proposed to use subsequently for the determination of ergosterol. TABLE I RIxo:!r)m EXTINCTIOX VALUES OF ERGOSTEROL AT 281-5 TO 282 mp (MAX.) Authors E Morton, Heilbron and Kamm* . . . . . . . . - 10,200 Morton and Gillam (quoted by Bacharach, Smith ancl Morton and de GouveialO . . . . .. . . - 11,700 Hogness, Sidwell and Zscheilell . . . . . . - 10,600 Huber, Ewing and Kriger' . . . . .. . . - 11,500 Lamb, Mueller and Beach12 . . .. . . . . 289 12,000* * E calculated, assuming the monohydrate composition.Stevensons) . . . . . . .. . . . . 333 13,800*Sept., 19531 ERGOSTEROL IN YEAST. PART I 51 Z The chief impurities likely .lo be encountered in commercial ergosterol are zymosterol and 5-dihydroergosterol. Callow13 stated that the former, but not 5-dihydroergosterol, could be eliminated by recrystallisation from an alcohol - benzene mixture. By recrystallisa- tion of the benzoate and subsequent regeneration of the sterol he obtained a purified product, but unfortunately did not record its ultra-violet absorption values. Our experience supports the view that recrystallisation of the sterol does not invariably yield a pure product and that purification is most satisfactory after recrystallisation of a suitable ester.Ergosterol crystallises from hydrated solvents with one molecule of water of crystallisation (C,,H,,O.H,O requires 4-3 per cent.), but the moisture content as found by analysis normally differs a little from the theoretical figure, and it appears necessary to carry out this determina- tion on any material to be used as standard. To avoid any confusion, our results, unless otherwise stated, are expressed throughout in terms of anhydrous ergosterol. EXPERIMENTAL Spectrophotometers--The greater part of this work was carried out with ;i Uriicam spectro- photometer (model SP 500), the more important results being confirmed on a second Unicam or a Uvispek instrument. The wavelength scale was checked as a routine with the 486-1-mp hydrogen line and occasionally with the mercury lines at 313.2 mp.The maximum error in wavelength near the ergosterol maxima appeared to be about 0-2 mp. The optical density scale and cell thickness (1 cm) were checked by means of 0.006 per cent. w/v potassium dichromate in slightly acid solution, in the manner described by Cama, Collins and Morton,14 with whose figures ours were in close agreement (E::m equal to 124.6, 144.8, 48.8 and 106.8 ,-.,t 235 mp (niin.), 257 mp (max.), 323 mp (min.) and 350 mp (rnax.), respectively). The two main maxima on the ergosterol absorption curve (Fig. 1) are sharp and the band-width used has some effect on the measured optical density at the maxima for any given solution. The maximum permissible band-width does not have to be found when a spectrophotometer designed to be operated at a constant band-width of the order of 0.5 mp is used.Nevertheless, spectrophotometers requiring a constant energy level should be operated at the narrowest slit-width (and hence band-width) compatible with adequate electrical sensitivity of the instrument at the particular wavelength used. In practice, on a Unicam instrument, slit-widths up to 0.5 mm at 282 mp and 0.6 mm at 271.5 mp may be used without perceptibly lowering the observed optical densities. These slit-widths corre- spond, according to the manufacturer's formula, to nominal band-widths of 1.8 and 1-9 mp, respectively. To satisfy these requirements the hydrogen lamp must be of high emission at the wavelengths used and any solvent used must be of good transmittancy. SOLVENTS- Absolute alcohol-Usually the laboratory reagent grade of absolute alcohol is satisfactory without further treatment.cycloHexane-The commercial grade of cyclohexane usually contains traces of benzenoid impurities that cause a marked decrease in transmittancy below 282mp. As a rule these are insufficient to affect readings at the 282-mp ergosterol maximum, but, when additional readings are required at the 271.5-mp maximum, it is necessary to use cyclohexane containing as little as possible of these impurities. Alternatively, cyclohexane specially purified for spectroscopy can be med. Chloroform-Analytical reagent grade chloroform usually requires no further treatment. Ethylene dichloriae--Commercial ethylene dichloride can be freed from traces of acid and other impurities by shaking it twice with a small volume of 50 per cent.w/w aqueous potassium hydroxide. The aqueous washings are discarded and the solvent is filtered through a dry paper, dried with phosphorus pentoxide, filtered again and distilled; the first and last tenths are discarded. PREPARATION AND PURIFICATION OF ERGOSTEROL- Direct recrystallisation-The three methods generally used for the direct purification of ergosterol involve recrystallisation from a mixture of 1 part of 95 per cent. alcohol and 2 parts of benzene,' 95 per cent. alcohoP3 or acetone containing 1 per cent. of water.12 To test the efficiency of these methods of purification, 50-g portions of commercial ergosterol51 2 SHAW AND JEFFEIUES THE DETEIIMTNATION OF [Vol. 78 of good quality (E:Fm (anhyd.) at 282 mp = 204 in absolute alcohol) were recrystallised a number of times from each solvent.After each recrystallisation the ergosterol was dried in vacuo over calcium chloride and the moisture determined in a "pistol"-type apparatus with boiling toluene (b.p. 110" C) in the outer jacket. The ultra-violet absorption in absolute alcohol was determined after each recrystallisation ; tlir resdts are recorded in Table IT. TABLE I1 RECRYSTALLISATION OF ERGOSTEROL Number of recrystallisa- Solvent tion Alcohol (95 per cent.) -benzene 1 3 4 A 6 2 3 4 5 1 water 2 3 4 6 mixture ( 1 + 2 ) 2 Alcohol (95 per cent.) . . . . 1 Acetone containing 1 per cent. of Amount Iost at 105" C in vucuo, 5-0 4.9 4.8 4.9 5.0 4.9 4-5 4.5 4.2 4.6 4.8 4-8 3.7 4.9 4.9 4.9 Yo E i T m (calculated as anhydrous) 271.6 mp (max.) 283.0 283.3 284.5 285.1 287.2 285.8 279.5 280.3 279.2 282.7 282.8 280.1 281.4 283.0 284.3 282.5 282 m p (Inax.) 297.3 298.4 299.0 300.5 303.1 301.0 294.3 295.0 294.6 298.0 297.2 295.4 296.7 298.9 299.3 297.9 293.5 mp (max.) 169.5 169.7 170.3 170.4 172.3 171.1 167.5 167.9 167.f; 168.8 169.0 167.7 168.6 170.3 170.1 169.4 They show that alcohol - benzene appears to be the most satisfactory solvent, although acetone with 1 per cent.of water is only slightly less efficient. Extinction values, however, are 1 to 2 per cent. lower than those subsequently obtained after purification by means of the benzoate. PuriJiration through ergosteroZ benxoate-A 50-g portion of the same commercial sample as that used for the recrystallisations just recorded was benzoylatecl in dry pyridine, sub- stantially as described by Cal10w.l~ The resulting material (46 g) was recrystallised five times from ethyl acetate to give a product having physical constants (Table 111) close to those quoted by Callow.The ultra-violet absorption curve for the purified benzoate in chloroform is recorded in Fig. 1. TABLE I11 RECKYSTALLISATION OF ERGOSTEROL BENZOATE FROM ETHYL ACETATE E:F~ (max.) a t 284 t o 284.5 m p in chloroform 20 Recry stallisation Melting point, (a) 1146.1 row " C approx. 0.002% w/v I 169.8 to 170.4 - 66.5 - 87.5 2 170.8 t o 171.5 - 67.7 - 87.1 3 170.8 t o 171.3 - 67.8 - 87.5 4 170.8 t o 171.3 - 67.7 - 87.2 5 171-0 to 171.3 -- 68.0 - 88.0 255.0 259.0 257.4 257.5 258*9* * Erbm (max.) at 273.5 t o 274 m p = 250.5, EiZm (max.) at 296 mp = 156.The purified benzoate (10.4 g) was hydrolysed with 3 per cent. w/v alcoholic potassium hydroxide ; the regenerated sterol was washed with alcohol and water and then recrystallised once from 95 per cent. alcohol (yield 7 g). All recrystallisations and the hydrolysis were carried out in an atmosphere of nitrogen to prevent oxidation. The purified hydrated sterol was obtained as colourless crystals, m.p. 163-6 to 164.4" C, (a):' - 129.5" (- 136.0" as anhydrous), 165.6" (- 174.0" as anhydrous), loss in vacuo at 105" C, 4.8 per cent. (C28H440.H20 requires 4.3 per cent.). Portions of this material were stored in sealed nitrogen-filled ampoules kept in a refrigerator. The absorption curve in absolute alcohol is shown in Fig. 1 and the specificSept., 19531 ERGOSTEROL IN YEAST.PART I 513 and molecular extinction coefficients in the solvents mentioned above are recorded in Table IV. DISCUSSION OF misums The discrepancies among the extinction values of ergosterol found in the literature appear to have arisen through failure to prepare pure samples by direct recrystallisation and, to a lesser extent, through small differences in moisture content of the material after recrystallisation. By recrystallisation of commercial samples, notably from a mixture of alcohol and benzene, almost pure ergosterol can be obtained, but the most reliable method involves purification through a suitable ester such as the benzoate. Huber el nL7 prepared several niti-obenzoyl esters, but did not use them as means of purifying ergosterol. The extinction values in absolute alcohol for the hydrated sterol shown in Table IV are about 5 per cent.higher than those recorded by Huber et aL7 and very slightly higher than those of Lanib, Mueller and Beach,l2 with whose results we are in substantial agreement. For the exceptionally high result obtained in 1933 by Morton and Gillam, and quoted by Bacharach, Smith and Steven~on,~ we have no explanation to offer. TABLE I V MEAN EXTINCTION VALUES FOR I’UKIFIII) EIiGOSTEliOL Wavelengths Number of of maxima Solvent determinations ( 0-5 mp), m p Absolute alcohol . . 9 271.5 282.0 293.5 Chloroform . . . . d 274.5 284.5 296.0 Ethylene clichloride . . 4 273.5 283.5 295.5 cycloHexanc . . . . 4 271.5 282.0 294.0 7 E:?m of hydrated ergosterol* 276.3 291.1 ( & 0.34)t 166.9 240.7 260.2 158-3 258.9 274.7 159.0 268.6 284.7 162.4 E:Fm of anhydrous ergosterol 290 306 174 253 273 168 272 289 167 282 299 171 Molecular extinction coefficient, 11,500 12,150 6,900 10,000 10,850 6,600 10,800 10,450 6,650 11,200 11,900 6,750 € * Loss a t 105” C in vacuo, 4.8 per cent.Results for hydrated ergosterol are given to four figures, The remaining extinction values have bern rounded off to be consistent with the precision of as observed. the measurements. t Standard error of the niean. In view of these discrepancies we thought it advisable to have the main extinction values of oiir material checked by Professor Morton, who kindly undertook an independent cxamina- tion of a portion of the purified hydrated sterol.He reported a maximum at 282 mp, E:m of 291, in absolute alcohol and a maximum at 282.5mp, E:Fm of 286, in cyclohexane (not correctc!d for moisture). He also observes that EiTm at 281-5 mp must be taken as between 286 and 290 for the monohydrate of ergosterol, which would correspond wit> 299 to 303 for the anhydrous material, if the hydrate contains ergosterol and water in molar proportions. He adds that, in the Liverpool laboratories, the maximum absorption is generally found to be at 2816mp, compared with 28143mp estimated on our particular instrument. I t will be seen that agreement between the values obtained at Liverpool and by us for purified ergosterol is excellent. Confidcnce in the results is increased when it is remembered that agreement was also good for potassium dichromate solutions measured by different observers with photo-electric instruments oi different types. Any analyst proposing to use ultra-violet absorption methods for determining ergosterol should check the wavelength scale of his particular instrument by means of a suitable light source emitting in the region of 282 mp and check the optical density readings against dichromate in the manner described.Further calibration with specially purified ergosterol should not then be necessary. In certain instances an analyst may desire to work against a “standard preparation” of the purest ergosterol available. Provided that the examination of a reference standard such as dichromate does not reveal any instrumental errors and the extinction values of514 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol. 78 his material do not depart significantly from the values given in Table IV, he can safely use his own figures for analytical purposes. REFERENCES 1. Liebcrman, C., Ber., 1885, 18, 1803. 2. Rosenheim, O., Biochem. J., 1929, 23, 47. 3. Heilbron, I. M., and Spring, F. S., Ibid., 1930, 24, 133. 4. Bruckner, J., Biochem. Z . , 1934, 270, 346. 5. ~ , Ibid., 1934, 274, 465. 6. Pesez, M., and Herbain, M., Bull. Soc. Chiuz. Iivance, 1949, 5, 760. 7 . Huber, W., Ewing, G. W., and Kriger, J., J . Anlev. Chewa. Suc., 1845, 67, 609. 8. Morton, 3. A\., Heilbron, I. M., and Kamm, E. I)., J . Cheiit. SOL, 1927, 2000. 9. Bacharach, A. L., Smith, E. L., and Stevenson, S. G., A?zaZysl, 1933, 58, 128. 10. Morton, R. A., and de Gouveia, A. J. A., J . Chem. Soc., 1834, 911. 11. Hogncss, T. R., Siclwell, A. E., aiicl Zschcile, F. P., J . Bid. Chem., 1937, 120, 23'3. 12. Lamb, I!. W., Mueller, A, and Beach, G. W., Inn. Eng. Chem., Anal. Ed., 1946, 18, 187. 13. Callow, R. K., Biochewz. J . , 1931, 25, 79. 14. Cama, H. R., Collins, F. D., and Morton, I<. A., Ibid., 1951, 50, 48. GLAXO LABORATORIES LTD. GREENFORD, MIDDLESIX Uecettzbev 22nd, 1952

ISSN:0003-2654    

DOI:10.1039/AN953780509b

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

514 The SHAW AND JEFFERIES THE DETERMINATION OF Deterrnina tion Of Ergosterol in [Vol. 78 Yeast Part 11. Determination by Saponification and Ultra-violet Absorption Spectroscopy I ~ Y W. H. (:. SHAW AND ,I. 1’. JEFI’ERIES (Presented at the meetkg of the Society 011 Wednesduy, May 20th, 1953) Part of the ergosterol found in yeast occurs in a combiiied form and treatment to liberate the free sterol is a necessary preliminary to its determina- tion. Most published methods make use of alcoholic or aqueous alkali hydroxides for this purpose, and the variety of strengths prescribed suggested that further investigation was necessary. The conditions for the destruction of the yeast cells and maximum recovery of ergosterol have been studied. I t was found that boiling under reflux with alcoholic sodium or ~~otassiurn hydroxide did not brcak down the yeast sufficiently.Aqueous sodium hydroxide at strengths from 10 to 50 per cent. w/w gave low and erratic results, additional ergosterol being recovered by subsequent alcoholic saponification of the residue. Concentrations of potassium hydroxide above 20 per cent. w/w in aqueous solution proved satisfactory, the strength required depending on the length of boiling under I-eflux. For the method dcscribed, 40 per cent. w/w aqueous potassium hydroxide has been selected, as it allows considerable latitude in time of boiling and in concentration. The ergosterol in the extracted unsaponifiable matter is separated by digitonin precipitation from non-sterol substances showing irrelevant absorp- tion in the same region as ergosterol.This precipitation has been shown to be quantitative, the separated sterol digitonides yielding solutions spectro- scopically indistinguishable a t wavelengths above 260 mp from those of purified ergosterol. A METHOD for determining ergosterol in yeast based on ultra-violet absorption is unlikely to suffer interference from other sterols present in yeast ; 5-dihydroei-gosterol and zymosterol do not possess the conjugated di-ene system of ergostero11y2 and show no absorption in the ergosterol range. The minor sterols, ascosterol, fecosterol and epistero1,l are isomers of zymosterol and would be expected to behave similarly. Hence, provided the sterols of yeast can be separated quantitatively from matter possessing irrelevant absorption, the ultra-violet method should be specific for ergosterol.Separation of the sterols from yeast is normally carried out after destruction of the yeast cells by boiling alkalis. However, Bills3 proposed a method in which matter showingSept., 19531 ERGOSTEROL IN YEAST. PART I1 515 irrelevant absorption is removed by a preliminary treatment of the yeast with 75 per cent. methanol. After filtration, the yeast is dried under reduced pressure at 70" C and the ergo- sterol is then extracted from the dry powder with successive small amounts of a boiling mixture of alcohol and benzene (2 + 1). The filtered extracts are evaporated to dryness and the residue is dissolved in absolute alcohol for measurement of absorption at the 282-mp, maximum. This method was given an extended trial, but was not found satisfactory; it appeared that matter possessing irrelevant absorption at 282mp had not been entirely removed by the methanol treatment. In addition, extraction by alcohol and benzene was not efficient, even with an increased number of extractions for twice the prescribed period; additional ergosterol was still recovered from the extracted residues by applying the saponification procedure developed subsequently.At least a part of the ergosterol of yeast appears to be present in a combined form not readily extracted by solvents, and complete recovery of ergosterol can only be attained after breaking down the complex. This is associated with destruction of the yeastcells and is normally carried out by heating with aqueous or alcoholic alkali h y d r o ~ i d e s , ~ ~ ~ ~ although formic acid has been used for this purpose.* The variety of conditions proposed for saponification suggested that further investigation was necessary.EXPERIMENTAL Digitoniiz-Commercial samples of digitonin normally show little absorption in the ultra-violet region. The absorption of an 0-5 per cent. w/v alcoholic solution in a 1-cm cell of the particular sample used was negligible over the range 200 to 400mp. Yeast-Most of the work was carried out on samples of fresh bakers' yeast (Saccharomyces cerevisiae) of a strain relatively high in ergosterol. Each sample was mixed thoroughly upon receipt, transferred to a closed container and stored in a refrigerator. The ergosterol content of these samples was found to be constant for at least a fortnight, after which they were discarded.Recovery of ergosterol as digitonide-Digitonin has long been used as a precipitating agent for sterols and the weight of the complex formed provides an approximate means of determining total free sterol^.^ However, variations in composition of the complex preclude its use for accurate gravimetric purposes. Ergosterol digitonide is insoluble in ethyl alcohol of strengths between 80 and 95 per cent. v/v, but in absolute alcohol it is sufficiently soluble for spectroscopic purposes. Recovery of about 1-mg amounts of ergosterol was checked in the following manner. Two-millilitre portions of an alcoholic solution of ergosterol containing 0.456 mg per ml were treated in stoppered centrifuge tubes with 2.50 ml of 0.5 per cent.w/v digitonin solution in absolute or in 80 per cent. v/v alcohol and with sufficient alcohol or water or both to make up 5ml and to adjust the alcoholic strength to within the desired range. The tubes were warmed for a few minutes in a water-bath at 70" C and set aside overnight. Any precipitates were separated by centrifugation and decanting the supernatant liquid. The residues were dissolved in warm absolute alcohol and the cooled solutions diluted to 50 ml for spectroscopic examination. The percentage recoveries, recorded in Table I, show that precipitation is quantitative from 90 per cent. v/v alcohol, although the precise concentration is not critical, and also that 1 mg of ergosterol as digitonide is almost completely soluble in 5 ml of absolute alcohol.In addition, they confirm earlier findings4 that the extinction of ergosterol is unaffected by combination with digitonin. A full absorption curve was determined on a specimen of digitonide derived from yeast and the curve above 260mp was found to be indistinguishable from that of purified ergosterol. TABLE I RECOVERY OF ERGOSTEROL AFTER DIGITONIN PRECIPITATION Ethyl alcohol, % v/v . . .. 99 to 100 95-0 90.0 85.0 80.0 Ergosterol recovered, yo . . negligible 98.9 99.5 99.4 99.4 Efect of boiling alkalis on purijed ergosteroG-The effect of boiling aqueous or alcoholic sodium or potassium hydroxide solutions on purified ergosterol was ascertained in the following manner.5 16 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol. 78 Approximately 20-mg amounts of ergosterol, i.e., the amount normally derived from about 5 g of yeast cake, were treated with 10 ml of N alcoholic sodium or potassium hydroxide or with strong aqueous alkali diluted with the amount of water present in 5 g of sample.The mixtures were heated under reflux for 2 hours, cooled, diluted and extracted three times with ether (anaesthetic grade). The combined ether extracts were washed with water, and the ether was then removed by distillation. The residue was examined spectroscopically a t a suitable dilution in absolute alcohol. Ergosterol was recovered quantitatively and in spectroscopically pure condition whether 20 or 50 per cent. w/w aqueous or 1.0 N alcoholic potassium or sodium hydroxide was used (Table 11). TABLE I1 RECOVERY OF ERGOSTEROL AFTER ALKALI TREATMENT Recovery after treatment with Alkali treatment r > sodium potassium hydroxide, hydroxide, 70 70 N alcoholic .... .. .. 99.3 100.2 20 per cent. w/w aqueous . . .. 99.7 99.6 50 per cent. w/w aqueous . . .. 99.7 99.7 That similar conclusions apply in the presence of yeast solids was shown during later experiments, in which no loss of ergosterol was observed during prolonged periods of heating with alkalis of the strength required for quantitative liberation of the free sterol. EFFECT OF ALKALI CONCENTRATION AND OTHER CONDITIONS ON THE SAPONIFICATION Alcoholic solutions-Most authors appear to favour alcoholic solutions for the saponifica- tion, and extended trials were given to both sodium and potassium hydroxide, usually in concentrations close to 1.0 N .These experiments showed that alcoholic sodium hydroxide solution was unsatisfactory, as the yeast solids were largely insoluble and bumping during the heating under reflux was severe. Alcoholic potassium hydroxide solution was better in both respects and considerable attention was paid to the conditions prescribed by Castille and Ruppol.* Results by their method, however, gave values only about 75 per cent. of those attained by the aqueous saponification method described below. For example, two samples of yeast cake gave 0.29, 0.32 per cent. and 0.40, 0-44 per cent. by the respective methods. The possibility of increasing the strength of potassium hydroxide in alcoholic solution in order to give higher recovery is limited by solubility considerations.Aqueous solutions-In this series of determinations 5-g portions of fresh bakers’ yeast were heated under reflux with 10 ml of alkali for different lengths of time. After dilution with water the solutions were extracted with ether in the manner described above for pure ergosterol. The unsaponifiable matter was dissolved in absolute alcohol to 50ml and the sterols in a 2.5 or 5.0-ml aliquot were precipitated with an equal volume of 0-5 per cent. w/v digitonin solution in 80 per cent. v/v alcohol. The separated digitonide was finally dissolved in absolute alcohol to give a concentration suitable for spectroscopic examination. Initially, aqueous alkali at concentrations of below 20 per cent. w/v was used; in particular, 20 per cent. sodium hydroxide was studied in detail, as saponification periods of more than 1 hour yielded little additional ergosterol.However, no independent method was available to check the validity of the results. The procedure described above was there- fore modified so that the ergosterol was extracted after saponification from the undiluted aqueous solution; the residual aqueous layer was then diluted with alcohol to give an alkali concentration of about 1.0 N . After further heating under reflux, the solution was again extracted with ether, and any extracted ergosterol was separated as digitonide for spectroscopic determination. The results obtained in this way on two samples of yeast are shown in Table 111. They show that 20 per cent. aqueous sodium hydroxide (equivalent to about 15 per cent.w/v in the saponification mixture, owing to the water in the sample) does not liberate all the ergosterol in yeast and that an additional amount of ergosterol, usually 5 to 15 per cent. of the total, is obtained by subsequent alcoholic saponification. OF YEAST-Sept., 19531 ERGOSTEROL IN YEAST. PART rI 517 Attempts were made to carry out the aqueous and alcoholic saponifications successively, that is, without intervening extraction of the ergosterol liberated in aqueous solution, but results were invariably low and erratic (e.g., 0-23, 0.26 per cent. on sample A, Table 111). These differences were probably due to adsorption of ergosterol by the yeast solids precipitated on the addition of alcohol; they could not be redissolved by reasonable dilution with water.TABLE I11 RECOVERY OF ERGOSTEROL FROM YEAST BY SAPONIFICATION Amount obtained by saponification with 20 per cent. w/v sodium Yeast sample hydroxide, % Sample A .. .. .. .. 0.316 0.309 Sample B . . .. .. .. 0.289 0.292 Amount obtained by subsequent alcoholic saponification, Total, 0.031 0.347 0-038 0.347 0-032 0.321 0.02 1 0.313 % % The effects of increasing the alkali strength and of using potassium hydroxide in place of sodium hydroxide were then studied, whilst the additional alcoholic saponification was applied as a check on complete recovery in those tests of which the results are marked with an asterisk in Table IV. In those tests destruction of yeast solids during the aqueous saponification was so severe that the previously noted precipitation was not observed on the addition of alcohol.TABLE IV EFFECT OF ALKALI AND CONCENTRATION ON RECOVERY OF ERGOSTEROL IN YEAST Reflux period Saponification conditions f A \ (5 g of yeast + 10 ml of alkali) 4 hour 1 hour 19 hours 3 hours Sodium hydroxide, 20 per cent. w/v . . .. - 0-292 - - 0.289 Sodium hydroxide, 50 per cent. w/w . . . . 0.218 0.244 - 0-328 Potassium hydroxide, 50 per cent. w/w . . 0.333 0.359 0.359* 0.355 Potassium hydroxide to a concentration of 50 per 0.338* - 0-328 - 0.359* cent. w/w in the saponification mixture 0.325* 0.313 * Subsequent alcoholic saponification yielded no additional ergosterol. The results in Table IV show that recoveries of ergosterol are low and erratic when sodium hydroxide is used and confirm an earlier observation' that results are more satis- factory with potassium hydroxide.In order to establish the most satisfactory concentration of potassium hydroxide and the time required for saponification, additional tests were carried out. The results, shown in Table V, are compared with those for similar concentrations of sodium hydroxide, and they show that potassium hydroxide at concentrations of between 20 and 50 per cent. w/w can be used, according to the time allowed for saponification. A saponification time of 2 hours and a concentration of 40 per cent. w/w were selected, as they allow considerable margin in both time and concentration. Some destruction of ergosterol appears to occur when the strength of potassium hydroxide is raised to 50 per cent. w/w in the mixture, whereas with sodium hydroxide at all strengths results are low and variable.METHOD REAGENTS- Potassium hydroxide, 40 per cent. w/w-Dissolve 100 g of analytical reagent grade potassium hydroxide, containing not less than 85 per cent. w/w of potassium hydroxide, in 113 ml of water. Digitonin-Specially selected for low absorption in the range 270 to 310 mp. Digitonin solution, 0-5 per cent. w/v-Dissolve 0.5g of digitonin in 80 per cent, w/v ethyl alcohol to produce 1OOml. Filter if necessary.[Vol. 78 PROCEDURE- Weigh 5 & 0.01 g (Note 1) of fresh yeast into a 100-ml flask provided with a ground-glass connection. Add 10 ml of 40 per cent. w/w potassium hydroxide solution, avoiding con- tamination of the ground-glass connection. Attach an air condenser, lubricating the joint with a few drops of water, and boil gently for 2 hours (conveniently on a hot plate).Cool, transfer the contents of the flask to a 250-ml separator with the aid of 2 to 3 ml of water and extract successively with three 50-ml portions of “anaesthetic grade” ether, using the first portion to rinse the condenser and flask. Mix the ether extracts in a second 518 SHAW AND JEFFERIES: THE DETERMINATION OF TABLE V EFFECT OF ALKALI CONCENTRATION AND SAPONIFICATION TIME ON RECOVERY Results are given as percentage of the result obtained under the conditions recommended (40 per cent. w/w potassium hydroxide for 2 hours) OF ERGOSTEROL FROM YEAST Saponification time Alkali strength* (per cent. w/w) 10 20 30 40 50 50 in sap. mixture , Potassium hydroxide Sodium hydroxide h 7 f \ 1 hour 2 hours 3 hours w 1 hour 2 hours 3 hours 21 23 23 21 22 25 51 96 98 69 88 94 87 100 100 76 81 82 97 100 100 57 72 62 97 98 99 85 87 91 96 88 88 - - - * 5 g of yeast and 10 ml of alkali.separator and wash with two 25-ml portions of water. Discard the aqueous washings and filter the ether through a small plug of absorbent cotton-wool into a distillation flask, rinsing the separator and filter with three 10-ml portions of ether. Distil the ether and dry the residue in a gentle stream of air. Dissolve the residue in absolute alcohol by warming, cool and dilute with absolute alcohol to 50ml in a graduated flask. If a correction procedure is to be applied, dilute accurately a portion of this solution and measure its ultra-violet absorption at the required wavelengths.Usually 2.50 ml diluted to 50 ml in absolute alcohol is a satisfactory dilution. We return to this matter in our next paper.1° Mix together in a stoppered centrifuge tube a suitable aliquot (2-50m1, see Note 1) of the solution and when necessary dilute to 5-Om1 with absolute alcohol. Add an equal volume of digitonin solution, warm in a water-bath at about 70” C and set aside overnight. Centrifuge to clear the supernatant layer and then decant as much as possible, avoiding loss of precipitate. Suspend the precipitate in about 5 ml of absolute alcohol and transfer the suspension to a 50-ml graduated flask. Wash the centrifuge tube with more absolute alcohol to a volume of about 40ml. Warm to dissolve the precipitate, cool and dilute to volume with absolute alcohol.Measure the optical density (E) of the solution at 282mp (max.) in a l-cm cell, using the plain solvent in the blank cell. The percentage of anhydrous ergosterol in the sample is given by- E x 50 x 100 306 x 2-5 x 2 x weight taken. NOTE 1 T h e amounts of sample and alkali, and the dilutions prescribed, are satisfactory for yeast cake containing about 72 per cent. of water and 0.4 per cent. of anhydrous ergosterol. Samples containing a higher proportion of water should be dried to about 70 per cent. moisture ; alternatively, a stronger potassium hydroxide solution can be used to maintain the correct strength during saponification. In testing dried yeast, each gram of sample requires the addition of 4 ml of water and 10 ml of 40 per cent. w/w potassium hydroxide. NOTE 2-It is essential that the final solution be free from even the slightest trace of turbidity. If necessary, the solution should be filtered, loss by evaporation being avoided. As a check on clarity the optical density at 310mp should be less than 0.01 for a reading of 0.5 to 0.6 at the 282-mp maximum.Sept., 19531 ERGOSTEROL IN YEAST. PART I1 519 REFERENCES 1. Fieser, L. F., and Fieser, M., “Natural Products Related to Phenanthrene,” Reinhold Publishing 2. Smedley-Maclean, I., Biochem. J., 1928, 22, 22. 3. Bills, C. E., Massengale, 0. N., and Prickett, P. S., J . Bid. Chem., 1930, 87, 259. 4. Csstille, A., and Ruppol, E., Bull. Acad. Me’d. Belg., 1933, 13, 48. 5. Heiduschka, A., and Lindner, H., Hoppe-Seyl. Z., 1929, 181, 15. 6. Goering, K. J., U.S. Patent 2,395,115; Chem. Abstr., 1946, 40, 2930. 7. Giral, F., and Dupont, M. G., Rev. SOC. Mex. Historia Natural, 1944, 5, 173. 8. Pesez, M., and Herbain, M., Bull. SOC. Chim. France, 1949, 5, 760. 9. P h a u , H., and Hardy, Z., J . Pharm. Chim., 1929, 9, 145. 10. Shaw, W. M. C., and Jefferies, J. P., Analyst, 1953, 78, 519. Corp., New York, 1949. GLAXO LABORATORIES LTD. GREENFORD, MIDDLESEX December 22n4 1952

ISSN:0003-2654    

DOI:10.1039/AN9537800514

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

Sept., 19531 ERGOSTEROL IN YEAST. PART I1 519 The Determination of Ergosterol in Yeast Part 111. Corrections for Irrelevant Absorption in Solutions of Ergosterol BY W. H. C. SHAW AND J. P. JEFFERIES (Presented at the meeting of the Society on Wednesday, May 20th, 1953) The unsaponifiable matter of yeast normally contains substances that possess ultra-violet absorption in the same region as ergosterol, so that the extinction a t the ergosterol maximum (282 mp in absolute alcohol) cannot be used directly as a measure of the concentration of ergosterol present. Although the digitonin purification described in Part I1 may be used for separating ergosterol in a spectroscopically pure condition, considerable time can be saved by applying a correction to the gross absorption. Two such procedures have been studied: (i) a correction of the Morton and Stubbs type, based on the 282-mp maximum and two subsidiary wave- lengths at which the purified compound possesses six-sevenths of the maximum absorption, and (ii) a new four-point correction based on the three ergosterol maxima (271.5, 282 and 293-5 mp) with a fourth reading at 310 mp, at which ergosterol shows little absorption.Procedure (i) suffers from the disadvantage that the maxima in the absorption curve of ergosterol are sharp and observations at the subsidiary wavelengths must be taken on steep portions of the curve. In consequence small errors in measuring wavelengths have a large effect on the correction factor and any instrument used for this procedure must be calibrated with pure ergosterol. Procedure (ii) is free from this disadvantage, but requires a greater range of wavelengths and linearity of absorption must be assumed over it.Results by both correction procedures are compared with those by the full digitonin method. The degree of agreement is generally good, but depends on the shape and amount of irrelevant absorption, which has been determined for a number of different strains of yeasts. SOLUTIONS of ergosterol obtained after saponification of yeast are spectroscopically “impure” and therefore the absorption at the main maximum (282 mp) cannot be used directly as a measure of ergosterol content. The digitonin purification previously described1 is available for separating free ergosterol from matter possessing irrelevant absorption ; although this purification is efficient, it is time-consuming.The possibility of avoiding the digitonin precipitation by applying some form of correction to the gross absorption, provided this could be done without undue loss of accuracy, was therefore investigated. The well-known Morton and Stubbs2 correction for irrelevant absorption is based on measurements of absorption at a maximum on the curve and at two subsidiary wavelengths, usually one on either side of the maximum. If the irrelevant absorption is assumed to be linear over the three selected wavelengths, it is possible by a geometrical construction to520 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol. 78 correct for the irrelevant absorption at the maximum. If the maximum occurs at wave- length A,, and the two subsidiary wavelengths are A, and As, and the extinction values of the pure compound a t these wavelengths are respectively Ex,, Ex, and Ex,, the following general formula can be applied for calculating the corrected absorption at the maximum- E1k1k2(A3 - - E2k1k2(X3 - &) - E3k1k2(A1 - EA, (corrected) = k1@2 - A,) + k,k,(A, - A,) + k,@l - A,) where k , = ExJEx,, k , = Exl/Ex, and El, E, and E, are, respectively, the observed extinc- tions at wavelengths A1, A, and A,.If the two subsidiary wavelengths are taken at points where the absorptions of the pure compound are equal, i.e., where ExI/Ex, = Ex,/Ex, and hence k, = k , = I<, the following simpler expression can be used- 3 - - E2(A3 - - ( A (A2 - A,) + K(A3 - A,> + (A1 - - A3> Ex, (corrected) = K It is in the second form, or some derived form, that the correction procedure has been applied extensively to spectroscopic determinations of vitamin A.The subsidiary wave- lengths chosen are those at which the absorption of the pure compound is some convenient fraction, usually six-sevenths, of the maximum. In the same form and by use of the same fraction, the correction can be applied to spectroscopically impure solutions of ergosterol. The main ergosterol maximum at 282mp is, however, much sharper than the maximum of vitamin A, so that great care is required in nieasuring the absorptions, particularly at the subsidiary wavelengths, which are on steep portions of the absorption curve. Fig. 1 shows on an extended scale the absorption curves of a purified specimen of ergosterol in the region of the 282-mp maximum.On an instrument calibrated as described previously3 the main maximum was found at 281.8 mp and the subsidiary wavelengths for six-sevenths absorption at 278.3 and 284-3 mp with absolute alcohol as solvent. In cyclohexane the corresponding wavelengths were 282-2, 279.0 and 284.6 mp. It can be calculated from Fig. 1 that a wavelength error of +0-2mp in setting the instrument at both subsidiary wavelengths gives a correction about 8 per cent. and an error of +06mp a correction approximately 19 per cent. too high. The absorption at the maximum is much less affected by small errors in wavelength and, since it is sharply defined, any appreciable shift of the maximum is normally encountered only in grossly impure solutions.The results in Table I show that the Morton and Stubbs type of correction can be applied in this way with a fair measure of success. It appears essential, however, that apart from the usual check, the wavelength scale of any spectraphotometer used in this manner must be especially calibrated for purified ergosterol in the region of the maximum. The precise wavelengths found for six-sevenths of maximum absorption can then be used for this correction. In using the four-point correction described below, this careful calibration is unnecessary. The following formulae were used in calculating the results in Table I- E at 278.3 - 4632 E at 281.8 E at 284.3 E at 2S1.8' (i) in alcohol, F = 7 - 2.68 x E at 279-1 - 4.01 E at 282.2 E at 284.6 E at 282.2' (ii) in cyclohexane, F = 7 - 2-99 x F is the factor t o be applied to the gross extinction at the 282-mp maximum.FOUR-POINT CORRECTION- Consideration of the characteristic absorption curve of ergosterol suggested that a correction based on the three maxima should not require the careful wavelength calibration with pure ergosterol necessary for the usual Morton and Stubbs correction. In addition, experience showed that the absorption of an alcoholic ergosterol solution at 310 mp is a useful guide to purity, since the extinction of pure ergosterol at that wavelength is comparatively small (E::m = 0.8). Thus a direct indication of the irrelevant absorption is obtainable at that wavelength. From the four selected points on the ergosterol curve (at the maxima at 271-5, 282 and 293.5 mp and at 310 mp) three three-point corrections are obtained for E at 282 (corrected)Sept., 19531 ERGOSTEROL IN YEAST.PART I11 521 by substituting the appropriate values in the general Morton and Stubbs expression given above- (i) 4.288 x E at 282 - 2.242 x E at 271.5 - 2.046 x E at 293.5, (ii) 3.228 x E at 282 - 2.347 x E at 271-5 - 0.880 x E at 310, (iii) 26.62 x E at 282 - 45.16 x E at 293-5 - 18-55 x E at 310. A fourth correction can be obtained by eliminating the E at 282 reading between (iv) 6.232 x E at 293.5 - 2.675 x E at 271.5 - 3.559 x E at 310. Correction (iii), based on observations at 282, 293-5 and 310 mp, is unsatisfactory, because it so happens that these three points are almost in a straight line on the absorption expressions (i) and (ii), thus- Wavelength, mp Fig.1. -4bsorption curves for anhydrous ergosterol in the region of the 282-mp maximum. Curve A : in absolute alcohol; curve B: in cyclohexane. Concentration approximately 0.002 per cent. w/v. The wavelengths for s/,ths maximum absorption required for a Morton and Stubbs type of correction are indicated curve for pure ergosterol. In consequence, the coefficient of each term of the correction is unusually high. The remaining three corrections can be applied individually as a test for linear irrelevant absorption in the manner described by Gridgeman.* Correction (iv) is open to the criticism that it places undue reliance on the reading at the subsidiary maximum at 2936mp. A more balanced equation can be obtained by taking the mean of (i), (ii) and (iv) in the form of the following four-point correction- E at 282 (corrected) = 2.506 x E at 282 + 1-395 x E at 293.5 - 2.421 x E at 271.5 - 1.479 x E at 310.E at 282, E at 293.5, E at 271.5 and E at 310 are the observed extinction values of the particular solution (in absolute alcohol) at the three respective maxima and at 310mp. If a 1-cm cell is used, the percentage (w/v) of anhydrous ergosterol in the test solution is given by E at 282 (corrected)/306. The corresponding four-point correction (cyclohexane as solvent) is- E at 282 (corrected) = 2.513 x E a t 282 + 1.205 x E at 294 - 2.327 x E at 271.5 - 1.389 x E at 310, and the percentage of anhydrous ergosterol (w/v) in the solution under test is given by E at 282 (corrected)/299.522 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol.78 Table I shows a number of results obtained on various organisms by the full digitonin method previously described.l For comparison, those obtained by applying both correction procedures described above to solutions of the unsaponifiable matter before digitonin precipitation are included. TABLE I ERGOSTEROL CONTENT OF VARIOUS YEASTS A. B. c. D. E. F. G. H. J - K. L. M. N. 0. P. Sample Bakers’ yeast (Saccharomyces cerevisiae) Bakers’ yeast (S. cerevisiae) Bakers’ yeast (S. cerevisiae) . . . . Rhodotorula gracilis .. .. S. Carlsbergensis . . .. .. S. Carlsbergensis . . .. .. Brewers’ yeast (top fermentation) Brewers’ yeast (sample G after Brewers’ yeast (bottom fermenta- Brewers’ yeast (S. cerevisiae) .. S. italicus . . . . .. .. S. chevalieri . . . . .. .. washing) tion) Vacuum-dried yeast . . .. Roller-dried yeast (S. cerevisiae) . . Roller-dried yeast (S. cerevisiae) . . Ergosterol r A \ BY digitonin Total precipita- solids, tion, Y O % 28.0 0-441 0.430 0.435 0.443 28-0 0-420 0.414 26.5 0.363 0.358 11-5 28-0 28.5 21.8 19.1 26-8 20.5 31-8 61.2 - - - 0-0681 0.0680 0.523 0.525 1-389 0.051 1 0.0512 0.0538 0.0548 0.0350 0.0361 0.0636 0.0639 0.898 1.778 1.86 1.86 0-455 0-365 By Morton By and Stubbs four-point type correction, correction. Y O 0.435 0.429 0.430 0.439 0.420 0.413 0.359 0.352 0*360* 0-356* 0*358* 0.0728 0.0741 0.522 0.525 1.354 0.0589 0.0585 0-0568 0.0575 0.0386 0.0390 0.0669 0.0704 0.887 1.799 1-86 1.87 0-443 0.344 Yo 0.434 0.431 0.432 0.452 0.407 0.408 - - 0*343* 0*333* 0.324* 0.0645 0.0687 0.526 0.516 1.427 0.0534 0.0530 0.0564 0.0582 0.0402 0.0406 0.0660 0.0677 0.888 1.841 - - 0.478 0.364 Calculated t to the dry sample, 1.56 Y O 1.49 1.36 0.59 1-87 4-88 0.23 0.28 0-13 0.3 1 2-82 2.9 1 1-86 0.45 0.36 * cycloHexane as solvent.t By digitonin precipitation. IRRELEVANT ABSORPTION- The success of the kind of geometrical correction procedure considered depends on the validity of assuming linear irrelevant absorption. In vitamin-A analysis it is not possible to test this assumption experimentally6 and so the question remains a subject of controversy. With the digitonin treatment described earlier,l it is possible to separate the ergosterol quantitatively from yeast in a “spectroscopically pure” solution ; the irrelevant absorption in the mother liquors after digitonin precipitation can then be determined. Fig.2 shows the irrelevant absorption determined in this way on certain of the samples listed in Table I. When the proportion of ergosterol in the dry matter is high, as in samples F and M, the relative amount of irrelevant absorption is small and little correction is required. Bakers’ yeasts Q, R and S, containing less ergosterol, gave satisfactory corrected results because the irrelevant absorption , although higher, was nearly linear. Brewers’ yeasts J, G and K, and the Rhodotomla strain, D, contained only small amounts of ergosterol, so that not only was the irrelevant absorption a much higher proportion of the gross extinction, but it also showed marked deviation from linearity.For such materials no geometrical adjustment will give more than an approximation to the correct value; for accurate results the full digitonin procedure must be used. The samples of brewers’ yeast contained a high proportion of fermentable matter, which appeared to be the source of some of the irrelevantSept., 19531 ERGOSTEROL I N YEAST. PART 111 523 absorption. This is illustrated by sample G, which gave about half the original irrelevant absorption (curve H) after being washed with water. This procedure should always be applied to yeast containing little ergosterol if it is desired to obtain an approximate result by the correction procedure for estimating ergosterol content. Wavelength, mp Fig. 2. Irrelevant absorption of yeasts expressed as a percentage of the ergosterol present. For key to samples, see Table I. Samples Q, R and S are representative bakers’ yeasts (Sacchuromyces cerevisiae) containing, respectively, 1.07, 1.42 and 1.60 per cent. of ergosterol in the dry matter 5 REFERENCES 1. 2. 3. 4. 5 . NOTE-References 1 and 3 are to parts I1 and I of this series. GREENFORD, MIDDLESEX Shaw, W. H. C., and Jefferies, J. P., Analyst, 1953, 78, 514. Morton, R. A., and Stubbs, A. L., Ibid., 1946, 71, 348. Shaw, W. H. C., and Jefferies, J. P., Ibid., 1953, 78, 509. Gridgeman, N. T., Ibid., 1951, 76, 449. Bagnall, H. H., and Stock, F. G., J . Pharm. Pharmacol., 1952, 4, 81. GLAXO LABORATORIES LTD. December 22nd, 1952

ISSN:0003-2654    

DOI:10.1039/AN9537800519

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

524 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol. 78 The Determination of Ergosterol in Yeast Part IV. A Short Method Based on Ultra-violet Absorption BY W. H. C. SHAW AND J. P. JEFFERIES (Presented at the meeting of the Society o n Wednesday, May 20th, 1963) An accurate method for determining ergosterol in yeast has been described previously; it involves saponification, extraction of unsaponifiable matter, digitonin precipitation and subsequent spectroscopic determination and has proved too lengthy for the examination of large numbers of samples. For this purpose, a short semi-micro method is described; it involves the saponification of a portion of the yeast sample containing about 1 mg of ergosterol with a small volume of 40 per cent. w/w aqueous potassium hydroxide in a specially designed flask and extraction of the unsaponifiable matter with a single accurately measured volume of cyclohexane.The filtered extract is suitable for spectroscopic examination. The four-point correction procedure, also previously described, is applied to allow for irrelevant absorption ; the ergosterol content of the sample is calculated from the corrected extinction a t the 282-mp maximum. The method requires a shorter manipulation time than the full digitonin procedure, but, because a geometrical correction and a single extraction only are used, gives results slightly lower and a standard deviation of a single determination a little greater than those got with the full method. IN a previous paper,l a method was described of determining ergosterol in yeast.Although it is accurate and therefore of special use for research purposes, the time necessary for a single determination by it is considerable because of the manipulations involved in saponifica- tion, three extractions with ether, distillation of the ether, precipitation with digitonin and spectroscopic examination of the digitonide in solution. If any irrelevant absorption present is assumed to be linear, the digitonin precipitation may be omitted2 with some saving of time, but not sufficient for an examination of many small samples of yeast, for which a short simple method was required, even though it might entail some loss of accuracy. Ultra-violet absorption provides an accurate means of identifying and determining small amounts of ergosterol, so that, provided the ergosterol can be liberated and quantita- tively recovered, correspondingly small samples of yeast can be assayed.With this in mind, we have studied the possibility of saponifying about 0.2 g of fresh yeast in a small volume of aqueous potassium hydroxide and then extracting once with an immiscible solvent. EXPERIMENTAL EFFECT OF STRENGTH AND VOLUME OF POTASSIUM HYDROXIDE- Initially, saponifications were carried out on 0-2 g of yeast cake with 0.4 ml of 40 per cent. w/w aqueous potassium hydroxide solution, because it is desirable with a single solvent extraction to keep the volume of the aqueous phase as small as possible. Under these conditions ergosterol could generally be satisfactorily recovered by one extraction with 50 ml of solvent, but certain yeasts containing little ergosterol gave gelatinous residues from which the ergosterol could not be extracted quantitatively.Moreover, even with a specially-designed apparatus, it was difficult to avoid undue concentration of alkali and consequent destruction of some ergostero1.l About 1.5ml appeared to be the smallest volume in which the saponification could reasonably be carried out; a convenient apparatus for this and the subsequent extraction is shown in Fig. 1. The conditions necessary for the quantitative liberation of ergosterol from yeast have been investigated during the work already rep0rted.l However, it became apparent that with only one extraction the concentration of the potassium hydroxide solution had to be controlled over a narrower range than formerly (Table I).Sept., 19531 ERGOSTEROL I N YEAST.PART IV 525 The low results obtained when the concentration during extraction was much over 40 per cent. w/w appeared to be due to some adverse effect on the distribution of ergosterol between solvent and aqueous phase. Also, concentrations appreciably below 32 per cent. w/w led to low recovery, possibly on account of incomplete saponification. J - 824 ,20*5 (ext) 25 (ext) '(22 (int) Fig. 1. Pyrex-glass flask for saponification and extraction. Dimensions are in millimetres SOLVENTS- subsequent extraction. a few trials were carried out with it. Under the saponification conditions described various solvents were tried for the Ether-As diethyl ether is normally used in the determination of unsaponifiable matter, Although separations were rapid, the extracts were TABLE I EFFECT OF POTASSIUM HYDROXIDE CONCENTRATION ON RECOVERY OF ERGOSTEROL Strength of potassium hydroxide reagent, * 30 35 40 40 40 40 % w/w Concentration of potassium hydroxide during extraction, Maximum recovery, 27 82 32 100 36.5 100 99 97 417 86 46t 51t Yo w/w Y O * 1.5 ml with 0.2 g of yeast cake, saponified 2 hours.t Extra solid potassium hydroxide added after saponification. invariably turbid and loss by evaporation occurred during the necessary filtration. In addition, the ether appeared to dissolve some water and the potassium hydroxide extracted the preservative (probably hydroquinone) from the anaesthetic grade of ether used. These effects seriously altered the ultra-violet transmission of the solvent and specially prepared ether was necessary.Apart from this, quantitative manipulation of the ether at room temperatures above 25" C became almost impossible. isoPropyZ ether, b.9. 68" C-Spectroscopically pure isopropyl ether was not available commercially and purification proved troublesome. In addition, this solvent suffered from some of the defects of diethyl ether.526 SHAW AND JEFFERIES: THE DETERMINATION OF [Vol. 78 Ethylene chloride (1 :2-dichZoroethane), b.p. 84" C-Giral and Dupont3 recommended substituting this solvent for ether in the extraction of ergosterol. Although it gave satis- factory extraction, the material available had to be purified before it could be used for spectroscopy. After extraction with this solvent separation was slow, for the density of ethylene chloride is much nearer that of 40 per cent.w/w potassium hydroxide than is that of any other solvent considered. Further, the extracts had to be dried and filtered for spectroscopic examination. cycloHexane, b.p. 81" C-This proved to be the most satisfactory solvent. It yields extracts that can easily be clarified by filtration; alternatively, after being allowed to stand for sufficient time, the clear solvent can be decanted directly for examination. It is un- affected by shaking with 40 per cent. w/w potassium hydroxide solution and suitable batches of commercial material can be selected for use without further treatment. CORRECTION FOR IRRELEVANT ABSORPTION- The four-point and other geometrical corrections2 depend for their validity on the assumption that the irrelevant absorption is linear over the selected wavelengths.For many samples of washed yeast containing much ergosterol, this assumption has been shown2 to be valid, and results obtained by the correction procedure are reliable. When, however, it is desired to determine ergosterol in yeast from a strain not previously examined, it seems essential to apply the full digitonin method previously described1 and to check by direct observation on the digitonin mother liquors that the irrelevant absorption is indeed linear over the wavelengths used. TABLE I1 COMPARISON OF ERGOSTEROL DETERMINATIONS BY THE DIGITONIN AND SHORT METHODS I BY Yo % Total digitonin Sample solids, method, Bakers' yeast ( S . cerevisiae) . . . . 29.0 0.345 Bakers' yeast (S.cerevisiae) . . . . 26.5 0.397 Vacuum-dried yeast (S. cerevisiae) . . - 1-75 S. carlsbergensis .. .. . . 28-5 1.55 S. carlsbergensis .. .. . . 30.8 1.71 Anhydrous ergosterol A \ By short method r A -l No. of tions standard error, determina- Mean and its % 4 0.348 ( *0*0036) 15 0.394 ( 0.0016) 4 1-68 (f0.0104) 4 1.49 (f0.0112) 4 1.66 (&0.0087) Correction procedures are most seriously in error when a break in the irrelevant absorption occurs near the main ergosterol maximum at 282mp. In using the four-point correction2 with observations at the maxima (at 271.5, 282 and 294 mp) and at 310 mp, it is sometimes possible to detect irrelevant absorption of this type by direct inspection of the observed optical densities. For purified ergosterol in cyclohexane the ratio of E at 271.5 to E at 282 is 0.94.When the ratio is less than this, or when it approaches this value with a high reading at 310 mp, non-linear irrelevant absorption must be suspected and the full digitonin method will have to be applied if accurate results are wanted. The reading at 310 mp is a direct indication of the amount of irrelevant absorption at that wavelength, where the absorption due to ergosterol is small. RESULTS- In Table I1 figures for representative samples of yeast of good quality, obtained by the method described below, are shown in comparison with those obtained by the full digitonin method previously described.l Results for materials containing much ergosterol are repro- ducible and generally in good agreement with those given by the full method.Recovery is usually within the range 96 to 100 per cent. For samples containing little ergosterol, or when the irrelevant absorption shows any marked deviation from linearity, greater divergence than this must be expected. I t is clear that the values by the short method are, for the richer yeasts, significantly, but only slightly (under 10 per cent.), lower than those found by the digitonin method (mean of duplicate determinations); for the two poorer yeasts there is no real difference. Thc, standard error of a single observation calculated for all thirty-one analyses is &0.0130.Sept., 19531 ERGOSTEROL I N YEAST. PART IV 527 METHOD APPARATUS- Re@x$aslz (see Fig. 1)-The flask is made from a suitable length of l-inch diameter tubing and a 150-ml flask.The condenser should be of such diameter that the space between it and the side of the base of the flask is as narrow as possible, so that moisture does not condense in the upper portion of the flask when it is in use. The distance between the bottom of the condenser and the base of the flask should be within the limits specified, although the thickness of the slip oi paper placed in the ground-glass connection during the heating allows a small measure of control. Hot-@&-This is covered with a layer of sand about 0-5 cm thick. “ Wrist-action” shaking machine*-If necessary the length of two of the opposed arms of the shaker should be increased, e.g., from 10 to 20cm, so that the vertical movement of the flasks during shaking is about 1-5 cm.REAGENT- Potassium hydroxide solution, 40 per cent. w/w-Dissolve 100 g of analytical reagent grade potassium hydroxide, containing not less than 85 per cent. w/w of potassium hydroxide, in 113 ml of water. SOLVENT- air should be not more than 0.2 at and above 271.5mp. cycloHexaize-The optical density of a l-cm cell filled with the solvent measured against PROCEDURE- Weigh accurately into the base of the reflux flask sufficient of the sample to contain about 1 mg of ergosterol. Add 1-50 ml of potassium hydroxide solution, avoiding contamina- tion of the upper portion of the flask with either sample or alkali. Insert the condenser and place a slip of paper in the neck to provide a vent. Attach the condenser to a water supply and place the bottom of the flask in the sand on the hot-plate.Heat gently under reflux for 2 hours, swirling occasionally. Remove the flask from the hot-plate, disconnect the water supply and remove the slip of paper. Allow to cool, add 50 ml of cychhexane from a pipette, discard the water in the condenser, and fix the condenser in the flask by means of a short length of adhesive tape. Extract in an inverted position on the shaking machine for not less than 10 minutes. Allow the phases to separate for at least 15 minutes, and then filter the solvent through a pleated Whatman No. 42 filter-paper. Avoid loss by evaporation during filtration and reject the first few millilitres of filtrate. Determine the optical density of the filtrate in a l-cm cell, with plain solvent in the blank cell.Take observations at the maxima at 271.5, 282 and 294mp and at 310mp, obtaining the four readings, E at 271.5, E at 282, E at 294 and E at 310, respectively. Calculate the corrected extinction at 282 mp from the expression- E at 282 (corrected) = 2.513 x E at 282 + 1-205 x E at 294 - 2.327 x E at 271.5 - 1.389 x E at 310. The percentage of anhydrous ergosterol in the sample under test is then given by- E at 282 (corrected) x 50 299 x weight taken * The authors wish to express their thanks to Mr. J. L. Holmes for assistance with the practical work and to Mr. H. J. Bunker (Barclay Perkins & Co. Ltd.), Dr. J. I. Webb (A. Guinness Sons & Co. Ltd.) and Dr. J. S. Lowe for kindly supplying samples of yeast. REFERENCES 1. 2. - _- , Ibid., 1953, 78, 519. 3 . NOTE-References 1 and 2 are to parts I1 and 111 of this series.Shaw, W. H. C., and Jefferies, J. P., Analyst, 1963, 78, 514. Girai, F., and Dupont, M. G., Rev. S O ~ . Mex. Historia Natzwal, 1944, 5, 173. GLAXO LABORATORIES LTD. GREENFORD, MIDDLESEX Decembey 22nd, 1052 * A “Microid” shaker, modified as described, has been found satisfactory.528 HARRISON AND RAYMOND: THE DETERMINATION OF [Vol. 78 DISCUSSION ON THE FOREGOING FOUR PAPERS DR. R. E. STUCKEY asked what was the best method of determining the purity of commercial ergosterol samples and whether there was a great difference between the result determined on the 282-mp peak alone and that obtained by the authors’ four-point correction method. MR. SHAW said that the validity of using the extinction at the main maximum for calculating the percentage of ergosterol in commercial material depended on the impurities present. If these were only the minor sterols of yeast, which had no absorption in the ergosterol range, the observed extinction required no correction and would give the true percentage. If any non-sterol matter possessing absorption a t 282 mp was present, the four-point correction could be applied and should give a reliable result provided the irrelevant absorption was linear. Even if it was not linear, the four-point correction would disclose the presence of irrelevant absorption, and i t was then preferable to precipitate a portion of the sample with digitonin. The percentage of ergosterol could be calculated from the extinction of the digitonide a t 282 mp. In practice, optical rotation of commercial ergosterol was normally determined as well as ultra-violet absorption, but in their experience i t was generally impossible to correlate the two results, probably owing t o the presence of minor sterols, which contributed to the optical rotation but not to the ultra-violet absorption.

ISSN:0003-2654    

DOI:10.1039/AN9537800524

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

528 The HARRISON AND RAYMOND: THE DETERMINATION OF Determination of Microgram Amounts of [Vol. 78 Calcium BY G. E. HARRISON AND W. H. A. RAYMOND (Presented at the meeting of the Society on Wednesday, May 20th, 1953) A micro-method for the determination of calcium is described; it is based on the precipitation of calcium molybdate and the subsequent forma- tion of molybdenum thiocyanate, which is determined absorptiometrically. As little as 4 p g of calcium can be estimated to within *lo per cent. The method is especially useful for determining calcium in serum from young children or from small laboratory animals. A comparison of the results with those found by other methods is included. WE have recently had to determine the calcium in small samples of neutral fluid containing about 20pg of calcium to within &5 per cent.Calcium determinations by gravimetric,l volumetric,2 73 c o l ~ r i m e t r i c ~ ~ ~ ~ ~ and nephelometric' methods have been described, but in all of them the accuracy falls considerably for amounts of calcium less than 40pg. Titration of the supernatant fluid with the sodium salt of ethylenediaminetetra-acetic acid, with murexide as indicator,*q9 gave consistent results down to 50 to 1OOpg of calcium, but was incapable of giving the accuracy we required. The following method, based on the absorptio- metric estimation of a molybdenum thiocyanate,lOyll has been found to give reproducible results within &2 per cent. (0-5pg of calcium per sample). A special advantage of the method is that the presence of phosphate, ferric or ferrous iron, magnesium, copper or aluminium does not affect the calcium determination.The method can also be used for the estimation of strontium and barium together with lanthanum, yttrium and the rare earths, although, of course, in a mixture, it is impossible to distinguish between them. METHOD REAGENTS- from sodium molybdate of analytical reagent grade. on standing; this frees the molybdate solution from calcium. solution with 25ml of pyridine. Sodium molybdate - pyridine reagent-Make a 5 per cent. solution in 50 per cent. alcohol Filter off any precipitate that forms Mix 75 ml of the calcium-free Potassium thiocyanate, 5 per cent solution. Stannous chloride reageizt-Make a 0.8 per cent. solution in 4 N hydrochloric acid by This isoAmyl alcohol.Standard calcium solution-A solution containing about 2 mg of calcium per 100 ml. dissolving 0.5 g of tin in 4 N hydrochloric acid and making up to 100 ml with the acid. solution should be freshly prepared for use.Sept., 19531 MICROGRAM AMOUNTS OF CALCIUM 529 PROCEDURE- Put a known aliquot of the solution for assay containing up to 30pg of calcium in a 15-ml centrifuge tube and add 1 ml or more of the sodium molybdate reagent. Place the mixed solutions in a bath of water at 70" C for about 30 minutes. Allow the sample to cool and then centrifuge it at 2500 r.p.m. at 15 cm radius for 10 minutes. Discard the supernatant liquor and drain the tube on filter-paper. Wash the precipitate three times with 50 per cent. alcohol, draining the tube well after each washing.Run 2.5ml of the 5 per cent. potassium thiocyanate solution into a 50-ml separating funnel from a burette and add 5 ml of isoamyl alcohol from a second burette. Dissolve the calcium molybdate precipitate in the centrifuge tube in 4ml of hot stannous chloride reagent and pour the solution into the funnel. Wash the tube with 4 ml of 4 N hydrochloric acid and then with distilled water until the total volume of liquid in the funnel is about 20 ml. Stopper the funnel and gently invert it two or three times, and then set it aside to allow the alcoholic extract to rise to the top. Run off the aqueous layer and pour the coloured extract into a 10-ml calibrated flask. Return the aqueous solution to the funnel and again extract the molybdenum thiocyanate with a further 3 ml of isoamyl alcohol, repeating the inversion and separation.Add the second extract to the first and make the total volume up to 10 ml with isoainyl alcohol. Invert the flask a few times to mix the solution. CALIBRATION STANDARDS- Prepare a calibration graph by carrying out the procedure on a series of aliquots of the standard calcium solution; volumes of 0.2, 0.4, 0.6, 0.8 and 1.0 ml are convenient. As it is the molybdate ion that in fact is being estimated, confirmation of the completeness of the precipitations of calcium in this calibration can be made by using equivalent amounts of sodium molybdate. This was done in our calibration procedure and the results are shown in Table I. I t can be seen that there is close agreement between the absorptiometric readings, and we concluded that the precipitation and recovery of calcium as molybdate was complete.TABLE I ABSORPTIOMETER READINGS FOR VARIOUS AMOUNTS OF MILLIMOLAR CALCIUM SOLUTION 0.2 0.3 0-4 0-5 8 12 16 20 Volume Change in Mean taken, Calcium, wedge reading reading ml PE: 0.1 4 19-2* 21-0 19-5 19.3 19.0 19.1 35.3 36.2 35.7 35-6* 53.7 57.2 54.0 51.1 65-1 65.6 67.2 72.1 66-0 89.1* 88-6* 88-8 89.1 90.1 0.6 24 99.4 106.5 103.3 104.0 103-1 0.7 28 122.2 122.1 0.8 32 146.9 142.5 1452* 143.5 142-1 140.9 * Equivalent volumes of sodium molybdate reagent taken for these readings.530 HAICIiISON AND RAYMOND: THE DETERMINATION OF [Vol. 78 ABSORPTIOMETRY- The absorptiometer used was a conventional double-beam instrument in which a 6-volt 30-watt small tungsten-filament incandescent lamp was used as a single light source.The barrier-type selenium photo-cells connected in opposition were used with a portable reflecting galvanometer to register the difference in light intensity between the two beams. A 10-mm absorption cell, a colour filter (Ilford Blue No. 602) and an annular neutral wedge of density range 0 to 1 were inserted in each beam. One of the absorption cells was filled with distilled water and remained a standard of reference. The other cell was filled successively with isoamyl alcohol and the alcoholic extract of molybdenum thiocyanate. With one wedge pre-set at a suitable value, the other was so adjusted as to bring the galvanometer spot back to the scale zero. In this way the absorption of each solution was measured in terms of an angular displacement of one of the neutral wedges.This procedure had the advantage that the wedge readings were independent of the linearity of response of the photometric cells and of variations in the light source. RESULTS INORGANIC SOLUTIONS- The results of applying the procedure to aliquots of a standard solution containing 4-0 mg of calcium per 100 ml are shown in Table I. The angular change in wedge reading is a linear function of calcium concentration in the specimens, so that Beer’s law is obeyed over the range of the observations. From the results we conclude that 4 pg of calcium can be estimated to within &lo per cent. More- over, the colours are nearly stable over several hours. This stability is shown by the following wedge rotation readings for an alcoholic extract of molybdenum thiocyanate recorded at different times after separation of the alcoholic extract: after 0.3 hour, 87.6; after 1.5 hours, 88.4; after 20 hours, 86.8. A comparison of the molybdate, ethylenediaminetetra-acetic acid (EDTA) and oxalate methods of determining calcium is shown in Table 11, in which the volumes taken for each type of asasy are also shown.I t will be seen that although the molybdate and EDTA estimations agree well, the standard oxalate procedure gives low values when the sample contains 0.2 mg of calcium or less. TABLE I1 COMPARISON OF THE PROPOSED MOLYBDATE METHOD WITH OXALATE AND EDTA METHODS WHEN USED FOR DETERMINING CALCIUM I N STANDARD SOLUTIONS OF CALCIUM CHLORIDE Calcium found, millimoles per litre, by oxalate method, EDTA method, molybdate method, A f 7 Standard solution 10 ml of solution 5 ml of solution 1 ml of solution of calcium, taken taken taken millimoles per litre 0.5 0.42 0.53 0-50 0-3 0.32 0-30 0.2 0.16 0.22 0.21 ORGANIC SOLUTIONS- A specially useful application of this method is to the determination of serum calcium in young children and in small laboratory animals, for which the volume of the blood sample is strictly limited.The presence of protein in serum makes it necessary to oxidise these specimens by adding a few drops of nitric acid and ammonium nitrate and taking to dryness on a sand-bath at about 200°C. Thereafter the method of analysis is similar to that for inorganic solutions. We have investigated this application by taking as little as 0.05 ml of horse serum, for which the molybdate method gave a result of 17.2 mg of calcium per 100 ml.The same value was found on applying the molybdate method to 0-2 ml of the serum. The mean result of several determinations on the same serum by the ethylene- diaminetetra-acetic acid method, for which 0.5-ml samples were taken, was 17.0 mg per 100 ml.Sept., 19531 MICROGRAM AMOUNTS OF CALCIUM 531 CONCLUSIONS The estimation of calcium based on precipitation of calcium oxalate from an alkaline solution fails if the sample contains 0-2 mg of calcium or less, owing to the incomplete recovery of calcium at these levels. The ethylenediaminetetra-acetic acid method is both simple and rapid, especially for the biologically important determination of calcium in serum.Again, however, samples containing at least 50pg are required, and, if the phosphorus content is as high as 10 per cent., the estimation fails owing to inhibition of the end-point in the titration. On the other hand, the proposed molybdate method gives average recoveries of 97 per cent. when the sample contains 4 pg of calcium, and is unaffected by the presence of phosphorus. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES Vogel, A. I., “Textbook of Quantitative Inorganic Analysis,” Second Edition, Longmans, Green & Kolthoff, I. M., and Sandell, E. B., “Textbook of Quantitative Inorganic Analysis,’’ The Macmillan Salomon, J., Gabrio, B. W., and Smith, C. F., Arch. Biochem., 1946, 11, 433. Kultner, T., and Cohen, H. R., J . Biol. Chem., 1927, 75, 517.Roe, J. H., and Kahn, B. S., Ibid., 1929, 81, 1. Tsao, M. U., Ibid., 1952, 199, 251. Murayama, M., J. Lab. Clin. Med., 1948, 33, 906. Holtz, A. H., and Seekles, L., Nature, 1952, 169, 870. Debney, E. W., Ibid., 1952, 169, 1104. Braun, A. D., 2. anal. Chem., 1867, 6, 86. Hurd, L. G., and Allen, A. 0.. Ind. Eng. Chem., Anal. Ed., 1936, 7, 396. Co. Ltd., London, 1951, p. 587. Co., New York, 1947, p. 605. MEDICAL RESEARCH COUNCIL RADIOBIOLOGICAL RESEARCH UNIT ATOMIC ENERGY RESEARCH ESTABLISHMENT HARWELL, DIDCOT, BERKS. January 7th, 1963 D I sc u SSION MR. N. L. ALLPORT said that this was an interesting approach to the problem of determining calcium colorimetrically. Like other available methods it was an indirect one in that the substance determined was the molybdate ion attached to the calcium.In presenting the paper, Dr. Harrison had said nothing about the specificity of the procedure ; obviously lead, if present, would interfere, although, of course, one would not normally find lead in blood serum ; if there had been medication with bismuth, the possible effects of this metal would need to be considered. A minute trace of zinc might well be present in many samples, as this metal was very widely distributed in animal tissues, and Mr. Allport asked if the authors had considered whether there was any disturbance from this factor and whether they had ensured that phosphates did not interfere. DR. HARRISON, in reply, said that the specificity of the procedure was referred to in the first paragraph of the paper, but this had been omitted from the presentation.As Mr. Allport observed, lead was only of biological interest in exceptional circumstances, and no examination had been made of the interference from this ion or from that of bismuth. They had, however, found that ferrous, ferric, aluminium and phosphate ions were without effect even when present in large amounts, and also that zinc or magnesium present in amounts equivalent to that of the calcium in the sample did not affect the estimation. DR. J. H. HAMENCE asked what technique had been used for the determination of calcium by the oxalate method, and whether ethanol or acetone had been used in the initial precipitation and in the subsequent washing. Both these solvents had the effect of depressing the solubility of calcium oxalate in water and therefore of giving a better recovery when only microgram quantities of calcium were present, DR. HARRISON replied that they had not used ethanol or acetone in the oxalate precipitation or in washing the precipitate. The oxalate had been precipitated according to the method of Clark and Collip ( J . Bid. Chem., 1925, 63, 461), in which an excess of ammonium oxalate was added to an acidified solution of the calcium to be estimated and the acidity reduced to near the neutral point by dropwise addition of ammonium hydroxide, universal indicator being used. The precipitate was then heated and subsequently cooled for several hours. The precipitate was washed with a 2 per cent. ammonium hydroxide solution.

ISSN:0003-2654    

DOI:10.1039/AN9537800528

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

532 MIDDLETON AND STUCKEY: THE PREPARATION OF BIOLOGICAL [Vol. 78 The Preparation of Biological Material for the Determination of Trace Metals Part I. A Critical Review of Existing Procedures BY G. MIDDLETON AND R. E. STUCKEY A critical review is given of methods that have been used for the destruc- tion of organic matter of biological origin preparatory to the determination of trace metals, and of the objections and criticisms that have been made of these methods. THE ever-growing interest in trace metals, whether as elements essential to normal biological development or as a potential source of danger to health, has resulted in the publication of numerous papers dealing with methods for the determination of such traces mixed or combined with large quantities of organic matter.For this purpose it is generally essential to isolate the mineral constituents after degradation or complete oxidation of the organic matter, and in determinations of this kind the emphasis has naturally been placed more on the determination itself and the results attained than on the preliminary treatment of the material. Reports of such investigations are to be found in journals dealing with biology, medicine, public health, toxicology and food analysis, and are generally indexed under the names of the metals or elements in question and the material examined. As only a small proportion have anything new to say on the subject of the destruction of organic matter, the mere search for references to methods of destruction of organic matter in biological material is in itself tedious, and it does not appear that any attempt has been made to review the subject as a whole.In this paper an attempt is made to supply this deficiency, and there will be described in Part I1 (in the press) a method for the destruction of organic matter in biological tissues and products for which several advantages are claimed. THE DESTRUCTION OF ORGANIC MATTER The reactions involved in the dry or wet ashing of organic matter, whether of animal or vegetable origin, are obscure and have not been studied in detail. They will vary, not only according to the reagents and methods used, but with the nature of the substances concerned, and it is impossible to treat the question as a whole without taking this into account. The compounds that make up such material mostly belong to one of four main groups- Carbohydrates-Carbohydrates, main constituents of many vegetable products, give on heating a residue of carbon that burns away readily in air at a fairly low temperature.They are oxidised without difficulty by the usual methods of wet combustion. Fats-Fats are present, often in considerable amount, in both animal and vegetable tissues. On heating they are to a large extent lost by volatilisation, and the Association of Official Agricultural Chemists1 recommends “smoking off” fat at about 350” C as a preliminary to ashing at a higher temperature. Fats are difficult to destroy by wet combustion, partly because they sublime on to the cooler upper parts of the vessel, and it is often advantageous to carry out the earlier stages of heating in an open vessel to facilitate loss of fat by volatilisation.Proteins-Proteins may be present in large proportions in both animal and vegetable products. Certain animal proteins, apparently because of their high content of lysine, tyrosine and tryptophan, are among the most difficult substances to decompose by wet methods, while the difficulty of dry ashing may be due, at least in part, to their content of phosphates and other salts. Proteins decompose at a temperature that is generally below 350” C, and leave a black residue resembling carbon. It is, however, necessary to use a considerably higher temperature before the decomposition is complete. Other comfiounds-Among other compounds that may be present in, or form the main part of, biological tissues and products are phosphatides, urea, bile salts, nucleic acid, porphyrins, resins, waxes and sterols.Sept., 19531 MATERIAL FOR THE DETERMINATION OF TRACE METALS 533 Of the first three, and most important, groups mentioned above, carbohydrates are present chiefly in, and often make up the greater part of, vegetable tissues; proteins form the major constituent of animal tissues, meat products, nuts, and some seeds and fruits, and are generally associated with fats.It is not surprising to find that methods for destroying organic matter that work satis- factorily when applied to substances of one type become troublesome when used for materials having quite a different composition: for example, the destruction of organic matter of vegetable origin and consisting largely of carbohydrates (it?., nearly all vegetable tissues except seeds) by methods of dry ashing or wet oxidation generally offers no special problems or particular difficulty; on the other hand, animal tissues (organs, meat products, blood, shell-fish and the like) often give trouble, especially where it is necessary to take a large quantity of tissue in order to obtain sufficient of the trace element for determination.SCOPE OF THE REVIEW The large number of proposed methods or variations of methods for dry or wet ashing of organic matter is due partly to the need for adaptation to the special character (volatility or liability to form insoluble products) of particular trace elements, and partly to a realisation of the unsatisfactory nature of existing procedures when applied to particular substances.Animal tissues especially (e.g., liver or shrimps) offer practical difficulties in the application of existing methods. In the past insufficient attention has been given, in the application of different methods, to the nature of the material being destroyed; here this factor is given greater emphasis. The present paper does not attempt to give a complete account of the preliminary methods used for the destruction of organic matter, but is restricted to certain applications. Synthetic organic compounds are ignored, and volatile elements such as boron and mercury require procedures of such a special character that they are better treated separately. Methods used for arsenic have been included, but for this element the principles are usually similar to those used for less volatile elements.Spectrographic methods, owing to the small quantity of material required for a determination, are also considered as a special subject and have not been included. An account of methods that have been used for the destruction of organic matter of biological origin with the object of determining trace elements (with the limitations described in the preceding paragraph) is therefore given, together with objections that have been raised against them, and some account of our own experience in particular instances. These methods are grouped under a number of headings according to the general method or reagents used. DRY METHODS METHODS- Dry ashing is a method frequently used for the destruction of organic matter, and it is generally considered advisable that it should be carried out at as low a temperature as possible in order to reduce the possibility of loss of trace metals by volatilisation and by the formation of difficultly soluble compounds, as for example by reaction with silica.Whilst a number of workers have ashed materials directly at an indefinite temperature-presumably a more or less bright red heat-others are more specific. A temperature of about 500" to 550" C (equal to a dull red heat), which represents the lowest temperature at which combustion can generally be completed in a reasonable time, is most commonly recommended, especially when the trace metal in question is volatile. For the determination of lead in animal tissues and similar material, Tiairhall2 recommends ashing at a temperature well below red heat, otherwise there is a loss by volatilisation; Schmidt3 first moistens the sample with sulphuric acid and then ashes at a dull red heat; Kehoe, Edgar, Thamann and Sanders4 use a temperature not over 600" C for ashing deposits from urine, as they find Fairhall's method unsatisfactory. Seiser, Necke and Muller5 also disagree with Fairhall and consider that blood can only be ashed without loss of lead if sulphuric acid is first added-they recommend a temperature of 500" to 530" C, and state that lead sulphate is not volatile below 550" C.TannahilP ashes the sample at a dull red heat when determining lead in tissues. Roche Lynch, Slater and Osler7 consider that when ashing bone, a temperature above 550" C will result in loss of lead, and that it is possible that prolonged heating at even a lower temperature may result in some loss.Weyrauch and Miillera used a temperature of 500" to 550" C for the same purpose.534 MIDDLETON AND STUCKEY: THE PREPARATION OF BIOLOGICAL [Vol. 78 For zinc, dry ashing is used by EggletonJ9 the temperature not exceeding 450” C, and by Rauschning,lo at 500” C; Heller and Burkell use a “black heat,” and Sylvester and Hughes,l2 followed by Allport and Moon,13 find dry ashing at 500” to 550” C satisfactory, with no appreciable volatilisation below 900” C. Forbes and Swift,15 determining iron in meat products, ignite the sample “to a white ash” at an unspecified temperature ; while Davies16 determines copper and iron in milk and cheese by ashing the sample over a smoky flame in a wide tube as an improvised muffle furnace.In the procedure used by Ansbacher, Remington and Culpl7 for copper in eggs, oysters and so on, the temperature of ashing does not exceed 400” C, although repeated moistening with nitric acid and re-ashing may be necessary to complete the procedure. The procedure of McFarlane18 for copper in blood is practically identical. Bertrand and Macheboeuflg ash the sample in a thin layer at a low temperature (unspecified) for cobalt and nickel in animal organs. Warburg20 ashes serum and similar material in a special silica flask at a dark red heat for determining copper, iron and manganese. Sylvester and Lampitt21 ash milk with a small quantity of sulphuric acid in a silica dish at 500” to 550” C for the determination of copper. With aluminium there is some danger of the metal becoming insoluble or of forming silicates if too high a temperature is used for ashing.Eveleth and Myers22 use a temperature not over 500” C for tissues, milk and urine, usually for a period of 24 hours, but it is sometimes necessary to digest the ash with sulphuric and nitric acids to remove the last remnants of carbon. For tin, which also is liable to become insoluble, dry ashing methods are not usually used, but M i ~ k ~ ~ ashed tissues with the addition of magnesium nitrate, and B a m f ~ r d ~ ~ similarly used a mixture of magnesium oxide and nitrate for determining antimony and arsenic, following the method of Str~yowski.~~ In order to determine silica in tissues, King26 fused the material with sodium carbonate in a platinum crucible-here, however, the quantity of material taken for the analysis was quite small.For determining cobalt in liver, Kidson, Askew and Dixon2’ found all methods of ashing with additions, whether of calcium acetate, calcium hydroxide , magnesium acetate or magnesium nitrate , to be unsatisfactory. Often no attempt is made to burn all the carbon in a single operation. TannahilP emphasises that tissues form a residue that must be repeatedly extracted before the entire char can be consumed, and that usually re-ashing is essential owing to fused inorganic salts preventing complete oxidation. Fairhal12 and McHargue28 extract the char with hydrochloric acid; Daviesl6 extracts the ash of butter with water and re-ashes the residue; other workers complete the removal of carbon by treating the residue with nitric and sulphuric acids in a wet combustion procedure (Seiser, Necke and Miiller5; Eveleth and Myerszg; Richards1*). Alternatively, re-ignition after moistening with nitric acid is sometimes used (DavieP ; Tompsett and Anderson29 ; Glassmann and Barsutzkaja30). Essery3l prefers to reduce the amount of sugars in wort by fermentation before ashing, followed by extraction of the residue with hydrochloric acid and re-ashing. A method intermediate between dry and wet ashing is that of Wirthle32-digestion with sulphuric acid until a porous mass is formed, followed by the addition of soda and sodium nitrate and ashing.Finally, mention should be made of the procedure of combustion in a closed vessel (bomb) in oxygen, used for the determination of arsenic by Bertrand,33 by Remington, Coulson and Kolnitz,34 and by Carey, Blodgett and Satterle~.~~ Among official methods for the determination of trace metals in biological products, those for gelatin in the British Pharmacopoeia, 1953,36 and the United States Pharmacopoeia, 14th edition,37 all recommend ashing at an unspecified temperature.The Dutch Pharmacopoeia, 1926,38 recommends a sulphated ash method. The British Standard meth0d3~ for gelatin is very indefinite-the material is incinerated in a platinum capsule “until a white ash is obtained’’ at a temperature that is specified as not high enough to cause loss of nietals by volatilisation and not so low as to produce errors due to colouring of the resulting solutions with “carbon”; the alternative of wet ashing (with nitric and sulphuric acids) is permitted.For the determination of lead, the Association of Official Agricultural Chemists recommends ashing food products40 at a temperature of 500” C with the addition of a proportion of aluminium and calcium nitrates; further portions of these salts or of nitric acid may be added if necessary. Wet combustion with nitric and sulphuric acids is recommended for a number of other elernent~.~1 Richards14 ashes at a “low red” heat for the determination of manganese.Sept., 19531 MATERIAL FOR THE DETERMINATION OF TRACE METALS 535 CRITICISMS AND OBJECTIONS- Various objections have been raised to dry ashing.These are based on the possibility of losing volatile elements, the absorption or combination of trace metals with other ash constituents or with the material of the vessel used, the possibility of contamination from dust, and the difficulty in carrying out the procedure. For the determination of cadmium in tissues, Cholak and H ~ b b a r d ~ ~ found that dry ashing was unsatisfactory whenever deflagration occurred and when samples containing more than 10 pg of cadmium were low in ash content. The addition of sodium phosphate resulted in a certain amount of improvement, but if too much was used the samples did not burn out completely and losses occurred when the ash was treated with nitric acid and re-ignited. Tompsett43 obtained poor recoveries of copper and iron from diets when ashed in silica vessels, but more satis- factory ones from urine and faeces, and concluded that the addition of sodium phosphate was to be recommended in respect of diets.Cheftel and Pigeaudg4 rejected dry ashing for the determination of lead in sardines as "we know by our previous experience that it leads to losses." With iron, the presence of chlorides gives rise to the danger of loss by volatiliFation, as ferric chloride is appreciably volatile at 450" C (Monier-Williamsg5), whilst with iron in complex combination the recoveries are variable and low (Woiwood46) ; Davidson47 found it necessary to remove silica with hydrofluoric acid before extracting iron from an ash. Sulphuric acid may be used to eliminate chlorides before ashing, and such a procedure was used for the determination of iron in biological material by Woiw0od,~6 but Weyrauch,48 who also used a sulphated ash at 500" to 550" C, considered the method to be unsuitable for organs owing to the difficulty of finding suitable vessels-phosphates in the ash may attack glass, quartz takes up oxides, and platinum might alloy with trace metals. Reimann and Minot49 also found the question of suitable vessels somewhat troublesome, since even when platinum is suitable, it is at times necessary to have a number of determinations in progress simultaneously.For determining manganese they used quartz beakers, specially selected for freedom from manganese, and after being ashed at 600" to 700" C, the partly burnt residue was treated with a mixture of sodium and potassium nitrates, sulphuric acid and hydrochloric acid, and then fused with the acid sulphates.C ~ m r i e , ~ ~ who treated milk by a sulphated ash procedure, noted a loss of copper resulting from contact with silica. In the process of dry ashing a good supply of air is necessary, and it is difficult to ensure the absence of dust containing, possibly, trace metals. Ansbacher, Remington and Culpf7 consider that the "dangers of loss and contamination in ashing in open dishes are not to be underrated." Woiwood46 also considers it necessary to take precautions against aerial contamination. Many workers reject the method of dry ashing completely. COMMENTS AND CONCLUSIONS- Although dry ashing-the most obvious method of destroying organic matter-would appear at first sight to be simple and reliable, in practice the simplicity is less apparent, because a first essential is the provision of special equipment in the form of a controlled- temperature mume furnace with fume disposal.The initial stage of carbonisation can be a troublesome process with animal tissues when large quantities are involved, as the outer layers harden and prevent the escape of steam and gases from the interior of the mass, and a voluminous froth results. It is better to dry the material first in an oven, but this is often very slow and does not completely eliminate the trouble. The next stage involves heating the dried mass in air at a temperature that is limited especially by the possibility of loss of trace elements by volatilisation and of formation of insoluble compounds such as silicates.The progress of the combustion varies according to the nature of the original material- possibly in part because of variations in the content of salts and of phosphorus-and with some animal tissues it is extremely difficult to burn all the carbon. With vegetable products, e.g., cocoa powder, the reverse is sometimes true-it is difficult to control the operation in such a way as to avoid incandescence. With a char derived from animal tissues, ashing at a temperature not exceeding a first visible red (k, about 480" C) is extremely slow, even in an electrically heated muffle furnace, and from the majority of writers quoted above it would appear that 500" to 650" C is required.536 MIDDLETON AND STUCKEY: THE PREPARATION OF BIOLOGICAL [Vol.78 Even at this temperature the process is slow or, with some materials, extremely slow, so that it is necessary to facilitate the combustion by removal of ash constituents, moistening with nitric acid, or by some other procedure. From this it can be seen that the application of dry ashing, especially to animal tissues, is often neither simple, rapid nor convenient, and that it does not necessarily lead to the recovery of the whole of the trace metals originally present. METHODS MAKING USE OF CHLORINE OXY-ACIDS (EXCEPT PERCHLORIC ACID) METHODS- Probably the earliest record of a method for the determination of a trace element in animal tissues is that of Fresenius and v. Babo51 for arsenic. They treated the organs with hydrochloric acid and potassium chlorate, filtering off and rejecting the fatty matter.Gusserow52 adopted a similar procedure, Friedmann53 modified the method by giving the sample a preliminary treatment for 24 hours with “antiformin” (a caustic alkaline hypochlorite solution) before adding hydrochloric acid and chlorate. The method of Kisskalt and F~iedrnann~~ is similar, and they showed also that the rejected fatty fraction generally con- tained only traces of lead. Schutz and Bernhardt,55 K ~ h n , ~ ~ and Eckardt5’ all used similar procedures, while Bernhardt58 used also euchlorine for decomposing bones. Harrold, Meek, Whitman and M ~ C o r d , ~ ~ determining indium in excreta and foods, boiled the material with nitric acid, then added concentrated hydrochloric acid and boiled to reduce the volume, and repeated this procedure twice more with more hydrochloric acid.Chloric acid was used for oxidation by Sonneiischein and Jeserich.60 The method of Fresenius and v. Babo is recommended by BamfordG1 as the best method for the detection of mercury in viscera and for the treatment of bones, although it is admitted that it has certain objections, namely, that it is long and tedious and that it involves the use of a number of reagents, some of which are difficult to obtain in a state of absolute purity. CRITICISMS AND OBJECTIONS- These methods are troublesome and unsatisfactory, and with a number of elements the, use of large quantities of chlorine compounds is undesirable on account of the danger of loss of trace elements by volatilisation.For example, Glassmann and Barsutzkaja30 found that tin was lost by volatilisation as chloride by this procedure. Further, the method at most breaks down organic compounds; it does not generally destroy them completely, and when insoluble matter remains after the treatment, e.g., with cellulosic materials, it has been shown that a large proportion of tin remains in this residue and can be recovered bjr fusion with alkali and nitrate (Manicka and Lauth62; D e ~ s s e n ~ ~ ) . A further disadvantage is that large quantities of salts are present in the final solution. DIRECT OXIDATION WITH PERCHLORIC ACID METHODS- For the determination of tin in canned foods, G O S S ~ ~ boiled the material down with perchloric acid, or with a mixture of nitric and perchloric acids.BolligeP5 also oxidised tissues by boiling with 60 per cent. perchloric acid and adding nitric acid drop by drop, followed by 30 per cent. hydrogen peroxide; the perchloric acid was removed by evaporation and any residue of ammonium perchlorate was destroyed by heat. This method was followed by Allcroft and Green66 for the determination of arsenic in tissues, and they suggest that “the use of sulphuric acid in place of nitric acid . . . is equally serviceable . . . when the quantity of organic matter is sufficiently small for the analyst to view a possible conflagration . . . with equanimity.” Hiscox67 found the perchloric acid method to give the best recovery of cobalt from plant materials. Gerntz6* also used nitric and perchloric acids, but without funiing off the excess of perchloric acid.CRITICISM AND OBJECTIONS- In spite of possible advantages of this method, few workers have felt disposed to adopt it, no doubt because of a fear of explosions. Those named above apparently succeeded in avoiding such mishaps, and probably the method is safe if the conditions laid down are carefully adhered to, but a momentary negligence might easily lead to a disaster. UnderSept., 19531 MATERIAL FOR THE DETERMINATION OF TKACE METALS 537 suitable conditions explosions resulting from perchloric acid and organic matter can be very violent. Having previously experienced such an incident when using perchloric acid with sulphuric acid for oxidising organic matter, we have not felt inclined to Eollow up the method described above, especially when somewhat large amounts of organic matter are to be destroyed.N’ET COMBUSTION WITH SULPHURIC ACID Methods of wet combustion with sulphuric acid can be considered as based on the original KjeldahP procedure for the determination of nitrogen, in which potassium permanganate was used as oxidising agent. For the determination of trace metals, a non-metallic oxidant is naturally preferred, and generally nitric acid, perchloric acid, or a mixture of both, is used. METHODS- The original procedure of Neumann for the determination of iron in proteins, urine and similar materials made use of a mixture of sulphuric acid and ammonium nitrate, but the procedure was soon simplified by replacing the latter by nitric acid.70 Necke, Schmidt and Klostermann7l used a mixture of sulphuric and fuming nitric acids for the destruction of organs and excreta, with the object of determining lead.Similar procedures were adopted by Schonheimer and F ~ s h i m a , ~ ~ Lampitt and Syl~ester,7~ Roche Lynch, Slater and O ~ l e r , ~ Hamen~e,7~ Klein and Wichmann,V6 Kidson, Askew and D i ~ o n , ~ ~ Wiihrer,77 Coleman and Gilbert,78 Adam and H0rner,7~ Lunde, Aschehoug and Kringstad80 and by Jones and Dawson.81 Cribb and Stills2 merely digested canned peas with nitric and sulphuric acids and filtered off undissolved matter, completing the wet combustion on the filtrate. Mi~iot,*~ determining arsenic in tissues, digested them with nitric acid overnight, boiled the digest for 2 hours, then added a few millilitres of sulphuric acid and completed the operation as a normal wet combustion. Lakemanna used a mixture of sulphuric and fuming nitric acids for the determination of arsenic, but found it necessary to complete the decom- position by fusion with potassium and sodium nitrates.Gross@ found that heterocyclic compounds remaining after the wet combustion of tobacco leaves with sulphuric and nitric acids interfered with the determination of arsenic, and that it was necessary to evaporate and ignite the digested solution. A modified procedure was adopted by Lampitt and Rookes6 for the determination of lead in canned sardines : the usual sulphuric - nitric acid treatment was continued until the liquid was straw-coloured, potassium sulphate was added and the heating was continued until the mixture was colourless; this took at least 4 hours.Other workers have endeavoured to complete the sulphuric - nitric acid combustion by the addition of perchloric acid (Bertrand87; Kehoe, Thamann and Cholaks8 ; Hubbardss; Morris and Calveryso ; CassiP ; Aull and Kinard92; Cholak and Hubbard42; Meunierg3; Feldstein and KlendshojS4). Kennedys6 and Myers, Mull and Morrison96 used sulphuric and perchloric acids only. McNaughtS7 found that, after the digestion of animal tissues with nitric and sulphuric acids, it was necessary to fume off the sulphuric acid and to heat the residue for 5 minutes at 500” C in order to destroy residual organic matter. Nitro-sulphuric acid combustion is specified in the British Standard method39 for gelatin for the determination of arsenic, and as an alternative method for other metals.Still others have dispensed with a special oxidising agent and relied on the oxidising action of sulphuric acid, the boiling point being raised by the addition of potassium sulphate. Such methods were used by de GiacomiSs for the determination of tin in minced meats, the use of nitric acid being considered inadvisable owing to the danger of forming insoluble tin compounds, by Willardg9 for copper in oysters, by Schryver100 for tin in canned meats, and by Chou and AdolphlOl for copper in urine and faeces. Catalysts as used for the determination of nitrogen have also been used, but are not favoured, as the action is slow and the most effective catalysts are metals or are liable to contain metallic impurities. Lawson and Scottlo2 used sulphuric acid with potassium sulphate and copper sulphate for determining arsenic in body tissues.Hilger and Labandelo3 used sulphuric acid with mercuric oxide when determining tin. In the method of Kleinlo4 for the determination of selenium, mercuric oxide is used with nitric and sulphuric acids, as it is claimed to “fix” the selenium and prevent its loss by volatilisation, but the method is not capable of destroying fats, which must be removed. This method has been adopted by the Association of Official Agricultural Chemists ,105MIDDLETON AND STUCKEY: THE PREPARATION OF BIOLOGICAL [Vol. 78 538 Nitrosyl-sulphuric acid was used by Francis, Harvey and BuchanlO6 to decompose the urea in urine as a preliminary to nitric acid treatment for the determination of lead.A reference to this reagent by Monier-William~~~7 suggests that this reagent has been used for the destruction of organic matter in general, but this does not appear to be so. CRITICISMS AND OBJECTIONS- Although carbon reduces sulphuric acid slowly at the boiling point, it is not practicable to oxidise animal matter with sulphuric acid alone, and even if a considerable quantity of sodium or potassium sulphate is added to raise the temperature, a number of heterocyclic compounds remain in the mixture undecomposed (Middleton and Stuckeylo8 ; Grosss5) , and the time required for the operation is extremely long when large quantities of organic matter are involved. The addition of catalysts such as mercury, selenium or copper is ineffective for dealing with large quantities of organic matter.In practice, therefore, sulphuric acid is nearly always used in conjunction with an oxidising agent, usually nitric acid. Although this procedure is generally convenient and satisfactory for vegetable products, it can be extremely troublesome and may even fail entirely when applied to animal tissues. Details of the procedure as adopted vary: nitric acid may be allowed to drop slowly and continuously into the boiling sulphuric acid digest, or the digest may be allowed to cool before the addition of nitric acid. In the first variant of the procedure the nitric acid is not used to best effect, as most of it immediately passes into vapour. Various comments have been made on the difficulty of carrying the digestion to a satis- factory conclusion (Kehoe , Thamann and Cholakss ; Necke , Schmidt and Klostermann71 ; Gortner and Lewislog ; Kleinl04; W o i w o ~ d ~ ~ ; Kidson, Askew and Dixon2’ ; Remington , Coulson and v.Kolnitz=; Cary, Blodgett and Saterlee35). Discussing the determination of copper in food- stuffs, Monier-Williams says that “if the sample contained much fat the acid residue should be heated with ammonium oxalate,”llO and “a slight yellow tinge . . . which is difficult to remove . . . appear to be caused by the action of nitric acid on fats with the production of substances which are extremely resistant. . . . Complete oxidation can usually be effected with perchloric acid. The yellow colour can sometimes be removed by heating with . . . ammonium oxalate. ”111 In practice , perchloric acid treatment is ineffective in difficult cases, while the purpose of the addition of ammonium oxalate is to remove nitrogen acids from the digest (Minots3), but neither perchloric acid nor ammonium oxalate can be relied on to give a colourless digest.In the determination of lead, precipitation of the lead as sulphate is one method that has been used for avoiding the difficulty caused by residual colour in the digest.l12 In our own experience, in the most difficult cases no combination of sulphuric acid with nitric acid or perchloric acid or both is effective in producing a colourless digest, and nothing is gained by the addition of hydrogen peroxide, ammonium oxalate or other similar oxidants. Peters113 considered the Neumann process of wet combustion inferior to dry ashing, which saves time, ensures the destruction of oxidising agents and gives a neutral ash.Gross86 and McNaught114 complain of the inconvenience resulting from the necessity of evaporating large quantities of sulphuric acid, and Wirthle32 also finds the large quantity of acid used in the so-called Kjeldahl procedure to be an unsatisfactory feature. With large quantities of different acids the blank may be appreciably higher than the amount of trace metal to be determined, and it is then necessary to distil the acids specially before use (Meunierg3; Roche Lynch, Slater and Osler7; Francis, Harvey and Buchanlo6; Lunde, Ashehoug and Kringstadso). This is a simple procedure with nitric acid, but inconvenient and undesirable with sulphuric acid or perchloric acid.USE OF NITRIC ACID ALONE The earliest use of nitric acid alone for the oxidation or destruction of organic matter is in the well-known method of Carius115 for the determination of sulphur, phosphorus, arsenic, and so on, in organic compounds. Temperatures of up to 320” C were used, with nitric acid of various concentrations up to that of the fuming acid, in a sealed tube. Although it is not clear whether Carius in fact applied his method to tissues, he suggests that it would be suitable for proteins, and also that it might be used for the determination of metals in organic combination.Sept., 19531 MATERIAL FOR THE DETERMINATION OF TRACE METALS 539 Much later, Fontes and Thivolle116 developed a procedure for the nitric acid combustion of tissues in which the use of sulphuric acid was avoided.The material (up to 50g) was treated in a Pyrex-glass Kjeldahl flask with its own weight of nitric acid and 1 or 2 ml of 10-volume hydrogen peroxide-the latter to oxidise nitrous vapours and to reduce frothing. After boiling the mixture for a short time to disperse the tissue, 1 g of magnesium oxide was added. The flask was then heated in a bath of fused metal at 300” C. When the excess of acid had been evaporated, the remaining porous mass charred and ignited. The residue was re-evaporated with a few millilitres of nitric acid, leaving a dry white ash readily soluble in acids. A procedure for the destruction of organs with pure nitric acid was recommended by Orfila, quoted by Danckwortt and Ude.l17 They digest organs in nitric acid for some hours (with fatty orpans the action is so vigorous that the flask must be cooled).Fat is removed and the liquid is evaporated to a dark syrupy mass, which is ashed in small portions over a small flame. The lead is extracted from the still carbonaceous residue with nitric acid, a s re-ashirig after moistening with nitric acid is stated to give rise to loss of lead by combination with silica. In a development of this method Danckwortt and Jui-gensllg digest bones for 14 days with nitric acid, and then heat in an acid-resistant cast-iron vessel. After the material has reached a syrupy stage there is a sudden “self-ignition” and a white ash is left. This operation was apparently carried out on successive small portions of the liquid.The authors claim to obtain a white ash without using a high temperature. The method of Cholakllg for bismuth in organs is somewhat similar, but the “self-ignition” is omitted. The samples are digested with nitric acid and the solution is evaporated and ashed in a muffle furnace at 450” to 500” C, the ash being moistened with nitric acid and re-heated. Because of difficulties due to the presence of phosphates, samples such as liver and brain were wet-ashed with sulphuric, nitric and perchloric acids. Kludt120 digested urine with nitric acid, evaporated the solution and heated the residue carefully. The mass showed localised incandescent spots (sparks), which should not spread through the mass. Faeces, treated similarly, gave a white ash without becoming incandescent.Vendevelde121 used a similar procedure for gingerbread, and McCance and Widdowson122 for faeces, urine and foodstuffs. Milton, Hoskins and J a ~ k m a n l ~ ~ used ammonium nitrate with nitric acid when deter- mining alkalies and alkaline earths (and sulphur) in diets and food products. About 1 g of dry material, dissolved in nitric acid, was treated in a 250-ml Kjeldahl flask with 10 ml of a 50 per cent. solution of ammonium nitrate in 25 per cent. nitric acid and heated. The procedure was repeated with more reagent as necessary. CRITICISM AND OBJECTIONS- The method of Carius1l6 is, of course, only suitable for small quantities of material, and even then is objectionable on account of the high pressures produced in the sealed tube. The methods both of Fontes and Thivolle116 and of Danckwortt et ~ 1 .~ ~ 7 ~ ~ are effective, but involve a stage in which the blackened residue from the evaporation of the nitric acid solution suddenly ignites and burns in the nitric acid vapour (assisted in the former method by the presence of magnesium nitrate). The temperature attained locally for a short period is obviously quite high, and it is to be expected that such a vigorous reaction might lead to loss of trace metals. It is, however, evident that this incandescence does not occur under all conditions, at least with some materials. Although these methods have the advantage of being simple and rapid, and give a soluble ash free from added reagents, they have not met with general acceptance. COMPARATIVE TESTS There are but few reports of comparative investigations of the suitability of various methods for a particular purpose.Allcroft and Green66 compared different methods of destruction of tissues for the determination of arsenic, and obtained the following recoveries : Kjeldahl digestion with potassium sulphate and copper, 20 to 63 per cent. ; nitric and sulphuric acids, 57 to 81 per cent.; potassium chlorate and hydrochloric acid, 0 to 49 per cent. ; perchloric and nitric acids, 63 to 97 per cent.; ashing with magnesium nitrate, 47 to 90 per cent. Hiscox67 compared six methods for the destruction of plant materials, with the object of determining cobalt: (1) by using nitric acid and ashing by method of Kidson and Askew;540 MIDDLETON AND STUCKEY: THE PREPARATION OF BIOLOGICAL [Vol.78 (2) using nitric and sulphuric acids; (3) by dry ashing; (4) by sulphuric acid digestion; (5) using perchloric and nitric acids; (6) by nitric, perchloric and sulphuric acid digestion, then taking to dryness. Jackson124 tried four methods of destruction for the determination of iron in biological material: (1) by dry ashing with the addition of sodium carbonate; (2) by sulphated ashing; (3) by dry ashing after covering with calcium carbonate ; (4) by wet ashing with nitric, sulphuric and perchloric acids. The test solution contained glucose, urea, sodium chloride and calcium phosphate. Method (1) was found to he unsatisfactory; method (2) gave a recovery of 40 to 54 per cent.; method (3) of 80 to 87 per cent.; and method (4) of 100 per cent.They concluded that wet ashing was demonstrated to be the only acceptable method of those tested. Manicka and LauthG2 compared the recovery of tin by various methods with the following recoveries: potassium chlorate and hydrochloric acid, 75 per cent. ; potassium chlorite and hydrochloric acid with hydrogen peroxide, 89 per cent. ; potassium chlorate and hydrochloric acid, followed by sodium hydroxide and potassium nitrate, 95 per cent. ; Wirthle method, 101 per cent.; Orfila method, 96 per cent.; sulphuric acid - nitric acid, 98 per cent.; sulphuric acid and hydrogen peroxide, 101 per cent.; sulphuric acid, hydrogen peroxide and nitric acid, 101 per cent. Conclusions drawn from the results quoted above can only be valid in their own limited sphere of application.The success of the subsequent treatment of the product after destruc- tion of the organic matter may depend on this preliminary treatment, and on the nature of the solution obtained. It is also necessary to take into account the chemical nature of the original material-whether of a protein, carbohydrate or fatty nature. They found the best recoveries of cobalt by method (5) or by dry ashing. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. REFERENCES “Ofiicial Methods of Analysis of the Association of Official Agricultural Chemists,” Seventh Fairhall, L. T., J . I n d . Hyg., 1922/3, 4, 9. Schmidt, P., Dtsch. Med. Wschr., 1928, 54, 520. Kehoe, R. A, Edgar, G., Thamann, F., and Sanders, L., J .Amer. Med. Ass., 1926, 87, 2081. Seiser, A., Necke, A., and Miiller, H., Z. mzgeeu. Chem., 1929, 42, 96. Tannahill, R. W., Med. J . Aust., 1929, 16, (l), 194. Roche Lynch, G., Slater, R. H., and Osler, T. G., Analyst, 1934, 59, 787. Weyrauch, F., and Muller, H., 2. Hyg. InfektKv., 1933, 115, 216. Eggleton, W. G. E., Biochem. J., 1939, 33, 403; 1940, 34, 991. Rauschning, S., 2. Untersuch. Lebensmitt., 1942, 84, 30. Heller, V. G., and Burke, A. D., J . Biol. Chem., 1927, 75, 85. Sylvester, N. D., and Hughes, E. B., Analyst, 1936, 61, 734. Allport, N. L., and Moon, C. D. B., Ibid., 1939, 64, 395. Richards, M. B., Ibid., 1930, 55, 554. Forbes, E. B., and Swift, R. W., J . Biol. Chem., 1926, 67, 516. Davies, W. L., J . Dairy Hes., 1932, 3, 86.Ansbacher, S., Remington, R. E., and Culp, F. B., Ind. Eng. Chem., Anal. Ed., 1931, 3, 314. McFarlane, W. D., Biochem. J., 1932, 26, 1022. Bertrand, G., and Macheboeuf, M., Bull. Soc. Chim. France, 1925, 37, 934. Warburg, A., Biochem. Z., 1927, 187, 255. Sylvester, N. D., and Lampitt, L. H., Analyst, 1935, 60, 376. Eveleth, D. F., and Myers, V. C., J . Biol. Chenz., 1936, 113, 449. Misk, E., Compt. Rend. Acad. Sci., Paris, 1923, 176, 138. Bamford, F., Analyst, 1934, 59, 101. Strzyowski, C., Pharm. Post, 1906, 39, 977. King, E. J., Biochem. J . , 1939, 33, 944. Kidson, E. B., Askew, H. O., and Dixon, J. K., N.Z. J . Sci. Tech., 1936/7, 18, 601. McHargue, J. S., J . Agric. Res., 1925, 30, 193. Tompsett, S. L., and Anderson, A. B., Biochem. J., 1935, 29, 1851.Glassmann, R., and Barsutzkaja, S., 2. Unlersuch. Lebensmitt., 1928, 56, 208. Essery, R. E., J . Inst. Brew., 1945, 51, 185. Wirthle, F., Chem. Ztg., 1900, 24, 263. Bertrand, G., Compt. Rend. Acad. ScZ’., Paris, 1903, 137, 266. Remington, R. E., Coulson, E. J., and v. Kolnitz, H., Ind. Eng. Chem., Anal. Ed., 1934, 6, 280. Carey, F. P., Blodgett, G., and Satterlee, H. S., Ibid., 1934, 6, 327. British Pharmacopoeia, 1953, p. 242. United States Pharmacopoeia, Fourteenth Revision, 1950, p. 257. Nederlandsche Pharmacopoeia, Fifth Edition, 1926, p. 220. British Standard Specification No. 757 : 1944. Edition, Washington, D.C., 1950, Chap. 24, 3 57.Seyt., 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67.68. 69. 70. 71. 72. 73. 74. 75. 76, 77. 78. 79. 80. 81. 32. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 19531 MATERIAL FOR THE DETERMINATION OF TRACE METALS 541 “Official Methods of Analysis of the Association of Official Agricultural Chemists,’’ Seventh Edition, Ibid., Chap. 24, 5 3. Cholak, J., and Hubbard, D. &I., Ind. Eng. Chem., Anal. Ed., 1944, 16, 333. Tompsett, S. L., Biochenz. J . , 1934, 28, 2088. Cheftel, H., and Pigeaud, M. L., A n n . Fnlsif., 1936, 29, 76. Monier-Williams, G. W., “Trace Elements in Food,” Chapman & Hall Ltd., London, 1949, p. 257. Woiwood, A. J., Biochern. ,I., 1947, 41, 39. Davidson, J., J. Ass. Og. Agric. Chenz., 1931, 14, 551. Weyrauch, F., Z. Hyg.InfehfKv., 1930, 1111, 1G2. Reimann, K., and Minot, A. S., J . Biol. Cheiw., 1920, 42, 329. Comrie, A. A. D., Analyst, 1935, 60, 532. Fresenius, R., and v. Babo, L., Ann. Chinz. Pharm., 1841, 49, 287. Gusserow, Virchow’s Archiv., 1861, 21, 443. Friedmann, A., Hoppe-Seyl. Z., 1914, 92, 46. Kisskalt, K., and Friedmann, A., 2. Hyg. InfektKv., 1914, 78, 500. Schutz, F., and Bernhardt, H., Ibid., 1925, 104, 441. Kuhn, B., 2. anal. Chenz., 1907, 46, 62. Eckardt, A., 2. UTztersuch. Nahr. -u. Genussm., 1909, 18, 193. Bernhardt, H., 2. anal. Chew., 1925/6, 67, 97. Harrold, G. C., Meek, S. F., Whitman, N., and McCord, C. P., J . Ind. Hyg., 1943, 25, 233. Quoted in Kobert, R., “Konzpendiuw dev praktischen Toxicologie,” Verlag von Ferdinand Enke, Bamford, F., “Poisons : Their Isolation and Identification,” Second Edition, J .(% A. Churchill Manicke, P., and Lauth, H., Plzami. Zentvnlh., 1927, 68, 161. Deussen, E., Arch. Phavm., 1926, 264, 360. Goss, R. C., J . I n d . Eng. Chenz., 1917, 9, 144. Bolliger, A., Aztst. J . Exp. Biol. Med. Sci., 1932, 10, 57. Allcroft, R., and Green, H. H., Biochem. J., 1935, 29, 824. Hiscox, D. J., Sci. Agric., 1947, 27, 136. Gerntz, H. W., I n d . Eng. Chem., Anal. Ed., 1935, 7, 167. Kjeldahl, J., 2. anal. Chein., 1883, 22, 3G6. Neumann, A., Hoppe-Seyl. Z., 1902/3, 37, 115; 1904/5, 43, 32. Necke, A., Schmidt, P., and Klostermann, M., Dtsch. wed. Wsclzv., 1926, 52, 1855. Schonheinier, R., and Oshima, F., Hoppe-Seyl. Z., 1929, 180, 249. Cox, H. E., Analyst, 1925, 50, 3. Lampitt, L. H., and Sylvester, N.D., Ibid., 1932, 57, 418. Hamence, J . H., Ibid., 1934, 59, 274. Klein, A. I<., and Wichmann, H. J., J . Ass. 08. Agric. Chem., 1945, 28, 257. Wuhrer, J., 2. Unfersuch. Lebensmitt., 1938, 76, 338. Coleman, D. R. K., and Gilbert, F. C., Analyst, 1939, 64, 726. Adam, W. B., and Horner, G., J . SOC. Chem. Ind., 1937, 56, 3 2 9 ~ . Lunde, G., Aschehoug, V., and Kringstad, H., Ibid., 1937, 56, 3 3 4 ~ . Jones, C. R., and Dawson, E. C., APzalyst, 1945, 70, 256. Cribb, C. H., and Still, A. L., Ibid., 1925, 50, 286. Minot, A. S., J . Cancer Res., 1926, 10, 293. Lakemann, G., Biochenz. J . , 1911, 35, 478. Gross, C. R., I n d . Eng. Chein., Anal. Ed., 1933, 5, 58. Lampitt, L. H., and Rooke, H. S., Analyst, 1933, 58, 733. Bertrand, D., Covnpt. Rend. Acad. Sci., Paris, 1942, 215, 590.Kehoe, R. A., Thamann, F., and Cholak, J., J . I n d . Hyg., 1933, 15, 257. Hubbard, D. M., I n d . Eng. Chem., Anal. Ed., 1941, 13, 915. Morris, H. J., and Cslvery, H. O., I n d . Eng. Chenz., Anal. Ed., 1937, 9, 447. Cassil, C. C., J . A s s . 08. Agric. Chenz., 1940, 23, 297. Aull, J. C., and Kinard, F. W., J . Biol. Chem., 1940, 135, 119. Meunier, P., Con@. Rend. Acad. Sci., Paris, 1936, 203, 891. Feldstein, M., and Klendshoj, N. C., Analyst, 1953, 78, 43. Kennedy, R. P., J . Biol. Chew., 1927, 74, 385. Myers, V. C., Mull, J. W., and Morrison, D. B., Ibid., 1928, 78, 595. McNaught, J. J., Analyst, 1939, 64, 23. de Giacomi, R., Ibid., 1940, 65, 216. Willard, J. T., J . Anzer. Chenz. SOC., 1908, 30, 902. Schryver, S. B., J . Hyg., 1909, 9, 253. Chou, T. P., and Adolph, W. H., Biochem. J., 1935, 29, 470. Lawson, W. E., and Scott, W. O., J . Biol. Chenz., 1925, 64, 23. Hilger, A., and Labande, L., 2. Untersuch. Nahr. -u. Genussm., 1899, 2, 795. Klein, A. K., J . Ass. 08. Agric. Chew., 1941, 24, 363. “Official Methods of Analysis of the Association of Official Agricultural Chemists,” Seventh Edition, Francis, A. G., Harvey, G. O., and Buchan, J. L., Analyst, 1929, 54, 725. Monier-Williams, G. W., op. cif., p. 84. Middleton, G., and Stuckey, R. E., J . Pharw. Pharnzacol., 1951, 3, 829. Gortner, R. A., and Lewis, H. B., I n d . Eng. Chenz., Anal. Ed., 1939, 11, 198. Washington, D.C., 1950, Chap. 24, 5 57. Stuttgart, 1912. Ltd., London, 1947. Washington, D.C., 1950, Chap. 24, 5 84.542 HOLDEN: A COMPARISON OF THE APPLICABILITY TO PLANT EXTRACTS [VOl. 78 110. Monier-Williams, G. W., op. cit., p. 51. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. -- , op. cit., p. 85. Peters, V. R. A., J . PhysioZ., 1912, 44, 131. McNaught, K. J., N.Z. J . Sci. Tech., 1938, 20, 1 4 ~ . Carius, L., Annalen, 1860, 116, 1; 1865, 136, 129; Ber., 1870, 3, 697. Fontes, G., and Thivolle, L., BzdZ. SOC. Chim. Biol., 1923, 5, 782. Danckwortt, P. W., and Ude, W., Arch. Pharm., 1926, 264, 712. Danckwortt, P. W., and Jiirgens, E., Ibid., 1028, 266, 492. Cholak, J., I n d . Eng. Chcm., Anal. Ed., 1937, 9, 26. Kludt, B., Hopfe-Seyl. Z., 1928, 174, 28. Vandervelde, A. J . J., 2. Untersuch. Nalzr. -u. Genussm., 1904, 7, 676. McCance, K. H., and Widdowson, E. M., J . PhysioZ., 1942, 101, 44. Milton, R., Hoslrins, J. L., and Jackman, W. H. F., Analyst, 1944, 69, 299. Jackson, S. H., Ind. Eng. Chew., A n d . Ed., 1938, 10, 302. -- , op. cit., p. 90. THE BRITISH DRUG HOUSES LIMITED GRAHAM STREET, LONDON, N . l April 27th, 1953

ISSN:0003-2654    

DOI:10.1039/AN9537800532

出版商:RSC    

年代:1953

数据来源: RSC

摘要:

542 HOLDEN: A COMPARISON OF THE APPLICABILITY TO PLANT EXTRACTS [VOl. 78 A Comparison of the Applicability to Plant Extracts of Three Methods of Determining Deoxyribonucleic Acid BY MARGARET HOLDEN The diphenylamine method (Dische, 1930) and a modification of the tryptophan - perchloric acid method of Cohen (1944) have been found to be satisfactory for determining deoxyribonucleic acid in extracts from plant tissue. The cysteine - sulphuric acid method (Stumpf, 1947) has given erratic results. A number of substances, some of which are likely to be present in plant extracts, have been tested under the conditions of the three methods. METHODS for the quantitative determination of deoxyribonucleic acid (DNA) have been used almost exclusively on animal tissues, although recently the diphenylamine method of Dischel has been used by Ogur and Rosen2 for DNA estimations on extracts from plant root tips and by McClendon3 on fractions from tobacco leaves.The diphenylamine method of Dische,l the cysteine - sulphuric acid method of Dische* and Stumpf5 and the tryptophan - perchloric acid method of Cohen6 have been investigated with a view to finding the most satisfactory one for determining DNA in extracts from plant tissues, with particular reference to the possible interference of pectic substances. Fructose and fructose derivatives, ascorbic acid and glyceraldehyde were found by Cohens to interfere in both the diphenylamine and tryptophan - perchloric acid methods. Fructose and its derivatives also interfered slightly in the cysteine - sulphuric acid method.6 MATERIAL AND METHODS Leaves from glass-house grown tobacco plants (Nicotiana tabacuum, variety White Burley) were used for most of the work. The leaves were minced in a domestic meat mincer andthe sap squeezed out by hand through strong cotton cloth.The fibre, i.e., the residue inthe cloth, which contains most of the DNA, was washed several times by suspending it in distilled water and squeezing. The washed fibre was stored at 4" C with chlorofonn added as a disinfectant. PREPARATION OF EXTRACTS CONTAINING DEOXYRIBONUCLEIC ACID- ethanol and ether to remove lipids. fibre in N perchloric acid at room temperature overnight (Holden'). by incubating the fibre in N perchloric acid at 37" C overnight. were usually almost colourless, but sometimes yellow or light brown.Fibre was first extracted with 0.2 N perchloric acid and then with a (3 + 1) mixture of Ribonucleic acid (RNA) was removed by soaking the The DNA was extracted Extracts made in this way Total carbohydrate was determined in the extracts by the orcin method of Pirie.*S o t . , 19531 OF THREE METHODS OF DETERMINING DEOXYRIBONUCLEIC ACID 543 Reducing sugar was determined by the Hanes9 modification of the Hagedorn and Jensen method. An equal volume of N sodium hydroxide solution was added to the sample to neutralise the N perchloric acid. Uronic acid was determined by Tracey’s method.lO The N perchloric acid extracts were neutralised with potassium hydroxide solution and after leaving them in a refrigerator overnight the precipitated potassium perchlorate was removed by centrifugation before drying the samples in bulb tubes.NitrogePt was determined by a micro-Kjeldahl method. Much of the perchloric acid was removed as potassium perchlorate, as large amounts of perchloric acid cause loss of nitrogen during incineration (Weeks and Frimingerll) . Absorption spectra in the visible region were measured with a Unicam diffraction-grating spec t rophot ome t er . DIPHENYLAMINB METHOD Three millilitres of solution containing DNA, or 3 ml of water for the blank, and 6 ml of Dische reagent (1 per cent. diphenylamine in a glacial acetic acid solution containing 2.75 ml of concentrated sulphuric The standard conditions used for the test were as follows. ”400 500 600 700 Wavelength, mp Fig. 1. Comparison of absorption curves of colours given by deoxyribose, DNA, a fibre Curve A, 10 mg of galacturonic acid; curve B, fibre extract; curve C, 2 mg of DNA; extract and galacturonic acid with the diphenylamine reagent.curve D, 200 pg of deoxyribose acid per 100 ml) were heated in glass-stoppered tubes in a bath of boiling water for 10 minutes. After cooling them in ice-water the solutions were transferred to matched tubes for measure- ment in an Evans Electroselenium photo-electric absorptiometer. The intensity of the blue colour given by deoxyribose and DNA was measured with the aid of Ilford No. 607 (maximum transmission 600 mp) or No. 626 (maximum transmission 570 mp) filters, the reagent blank being used for setting the zero reading of the instrument. The diphenylamine solution was made up from the AnalaR grade of reagent (supplied by Hopkin and Williams Ltd.), which gave a colourless solution, without recrystallisation, when dissolved in AnalaR grade of glacial acetic acid.Some samples of acetic acid contained a contaminant that gave coloured blanks and an “off-colour” blue with DNA when the liquor from a bottle containing partly crystallised acid was used. Interference could be avoided by using only the frozen portion. A dialysed solution of the sodium salt of thymus nucleic acid (supplied by British Drug Houses Ltd.) 2-Deoxy-L-ribose (from Prof. M. Stacey) was used for deoxyribose standards.544 HOLDEN: A COMPARISON OF THE AI-’PLICRBILITY TO PLANT EXTRACTS [VOl. ‘78 was used for DNA standards. There was a linear relationship between the intensity of colour and the amount of deoxyribose with standards containing from 15 to 130 pg and with DNA standards containing from 50 to 600 pg (3.5 to 42 pg of phosphorus). Solutions of wheat germ DNA (from Prof.M. Stacey) gave the same intensity of colour as thymus DNA solutions of the same phosphorus content. It is well known that purine-bound, but not pyrimidine- bound, deoxyribose gives a blue colour with the diphenylamine reagent. Hypoxanthine deoxyribonucleoside (from Prof. A. R. Todd) gave the theoretical amount of deoxyribose under the standard conditions of the test, whilst thymidine gave only a faint blue-green colour. Time of heating, minutes Fig. 2. Effect of time of heating with the diphenylamine reagent on the intensity of colour given by DNA, galacturonic acid, a fibre extract, DNA + galacturonic acid and fibre extract + galacturonic acid.Curve A, 500 pg of DNA + 1 mg of galacturonic acid; curve B, 1 ml of fibre extract + 1 mg of galacturonic acid; curve C, 500 pg of DNA; curve D, 1 ml of fibre extract; curve E, 1 mg of galacturonic acid The absorption curves in the visible region for the colours given by deoxyribose, DNA and a fibre extract, when heated with the diphenylamine reagent, are shown in Fig. 1. This shows that the absorption maxima are at 600mp, but that the fibre extract absorbs more strongly in the 400 to 450-mp region. Ogur and Roseii2 also found the maximum absorption to be at 600mp for DNA and plant tissue extracts. Deriaz, Stacey, Teece and Wiggins12 stated that 580 rnp was the wavelength of maximum absorption for 2-deoxypentoses when heated with diphenylamine.The difference may be due to the different type of instrument used. OTHER SUBSTAKCES GIVING COLOUliS WITH THE DIPHENYLAMINE REAGENT- GnZacturonic acid-Deriaz et aZ.12 tested the method for interference with a large number of substances, but did not include galacturonic acid. As pectin or its breakdown products are likely to be present in extracts from most plant tissues, pectin and galacturonic acid were tested under the standard conditions of the method. Pectin gave a gelatinous precipitate and a faint green colour when heated with diphenylamine. Galacturonic acid gave a blue colour different from that with DNA. The absorption curve shown in Fig. 1 indicates that there are maxima at 500 and 650mp.One to two millilitres of a fibre extract made with N perchloric acid was required for a DNA determination. The diphenylamine reagent was approximately 0.5 N with respect to sulphuric acid, so that the perchloric acid added significantly to the amount of strong acid present. The colour given by galacturonic acid was found to be decreased by decreasingSept., 19531 OF THREE METHODS OF DETERMINING DEOXYRIBONUCLEIC ACID 545 the concentration of strong acid. The diphenylamine was therefore dissolved in glacial acetic acid containing no added sulphuric acid. Without sulphuric acid diphenylamine solutions become coloured more quickly than did the usual reagent; a fresh solution had to be used for each batch. If a sample of less than 2 ml of a N perchloric acid extract was to be used for the test, N perchloric acid was added to ensure the presence of a total of 2 ml of it, and 2 ml of acid was added to the standards.The development of the colours given by DNA, galacturonic acid and DNA plus galact- uronic acid after different periods of heating with the diphenylamine reagent was investigated. The results with a No. 626 filter are shown in Fig. 2. The colour given by DNA developed quickly and reached a stable value, whereas the colour given by galacturonic acid did not develop at once, but developed slowly and was still increasing when the experiment was stopped after heating for 32 minutes. In other experiments with greater periods of heating Time of heating w-ith diphenylamine reagent, minutes Fig.3. Effect of degradation of pectin on intensity of colour with the diphenylamine reagent. Curve A, pectin solution, 4.6 mg per ml in N perchloric acid: 0.58 mg of reducing sugar; curve B, pectin after heating for 15 minutes: 0-93 mg of reducing sugar; curve C , after heating for 30 minutes: 1-46 mg of reducing sugar; curve D, after heating for 1 hour: 2.33 mg of reducing sugar: curve E, after heating for 2.3 hours: 3.40 mg of reducing sugar; curve F, after heating for 4 hours: 4.00 mg of reducing sugar the colour continued to develop, but the colour in the blanks also increased. On heating for only 4 minutes galacturonic acid caused no interference. Deriaz et aZ.12 heated for only 3-25 minutes, but as the colour given by DNA has not reached its maximum in this time it is advisable to continue heating until a stable value is reached. Filter No.607 has maximum transmission at the peak of the DNA curve, i.e., at 600 mp, but filter No. 626 with maximum transmission at 570mp gave lower values for galacturonic acid and was therefore used to diminish possible interference by galacturonic acid in the determination of DNA on fibre extracts. Fig. 2 also shows the results of heating l-ml samples of a fibre extract with the diphenyl- amine reagent for different times. With up to 16 minutes of heating the curve for the fibre extract ran parallel to the DNA curve, but for longer periods it gave more colour. Addition546 HOLDEN: A COMPARISON OF THE APPLICABILITY TO PLANT EXTRACTS [VOl. 78 of galacturonic acid caused marked divergence of the curves.The extract used in this experiment had a carbohydrate content of 2.6 mg per ml of which 0.8 mg per ml was reducing sugar and 1.4 mg per ml was uronic acid. Fibre extracts with higher carbohydrate contents showed greater divergence from the DNA curve, but this only occurred after heating for 8 minutes. A few extracts when heated with the diphenylamine reagent gave a gelatinous precipitate of pectic material that had to be removed before the absorptiometer reading was taken. The increase in colour with the diphenylamine reagent when pectin is degraded is shown in Fig. 3. Citrus pectin (by British Drug Houses Ltd.) was extracted with hot 70 per cent. ethanol to remove reducing substances and its uronic acid content was found to be 75 per cent.of the dry matter. The pectin was dissolved by heating to boiling in N perchloric acid to give a solution containing 4.6mg per ml. One millilitre of this solution was used for the initial graph of time of heating with diphenylamine. Its reducing sugar content expressed as galacturonic acid was determined and this was determined again after heating with perchloric acid for periods ranging from 15 minutes to 4 hours. The heating with acid was done in glass-stoppered tubes in a bath of boiling water. During the heating with acid, a white precipitate appeared in the solution and this was included in samples taken for heating with diphenylamine. The precipitate disappeared during the heating with diphenyl- amine in the samples that had previously been heated with perchloric acid for 1 hour or more.With samples that had been heated for less than 1 hour in perchloric acid a precipitate was still present after heating with diphenylamine ; this was removed before the optical density was measured. Yeast ribonucleic acid-This gave a faint green colour when heated with diphenylamine, but 5 mg caused an increase of less than 5 per cent. in the absorptiometer reading given by 200pg of DNA. Carbohydrates-Five milligrams of arabinose gave a faint green with diphenylamine, and 5mg of starch and glucose gave no detectable colour. Pirie8 found that agar, caragheen polysaccharides and aldehydo-sugars gave green colours with the diphenylamine reagent Two-milligram lots of agar, hepta-acetyl mannose, hepta-acetyl glucose, hepta-acetyl galactose, hexa-acetyl arabinose and hexa-acetyl rhamnose (all prepared by N.W. Pirie) were tested. Agar gave a bright green colour; the acetylated sugars gave only a faint green. Half of a milligram of 3 :6-anhydro-cc-methyl-~-galactoside (from Dr. D. J. Bell) gave a strong green colour and an absorption curve identical with that of 3 mg of agar. Digitoxose (by Hofmann la Roche), which is a 2 :6-dideoxyhexose, or methyl deoxypentose gave a yellow-brown colour of low intensity. Nitrogenous s~bstances-Davidson~~ stated that most colorimetric reactions for deter- mining sugars were prone to interference from proteins, and this has been confirmed for the diphenylamine method. Overend14 found that amino-acids and purine and pyrimidine bases increased the intensity of the colour given by DNA in this method, but this interference of amino-acids was not confirmed.* Fibre extracts made with N perchloric acid at 37" C (40 ml of acid were added to fibre containing about 1 g of dry matter) have a nitrogen content of less than 0.1 mg per ml so that little interference is likely to be caused.The precipitate did not adsorb colour from the solution. CYSTEINE - SULPHUKIC ACID METHOD One millilitre of sample con- taining DNA, or 1 ml of water for the blank, 0.1 ml of a solution of 5 per cent. w/v cysteine hydrochloride (by Roche Products Ltd.) and 10 ml of 70 per cent. w/w sulphuric acid were mixed and set aside at room temperature for 30 minutes for the pink colour to develop fully. The colour intensity was read with an Ilford No.623 filter (maximum transmission at 495 mp). There was a linear relationship between the amount of DNA and colour intensity for amounts of DNA between 150 and 8OOpg. For less than 150pg the relationship deviated from linearity, as found by S t ~ r n p f . ~ 2-Deoxy-L-ribose gave a more orange colour than DNA and the absorption curve (Fig. 4) showed strong absorption in the 400 to 430-mp region. With deoxyribose as standard, hypoxanthine deoxyriboside under the standard conditions of the method gave the theoretical amount of deoxyribose, whilst thymidine gave about 70 per cent. of the theoretical value. The standard conditions of the method were as follows. * Dr. W. G. Overend has informed me that he has since found that amino-acids did not increase the absorption due to DNA when other samples of diphenylamine were used.Sept., 19531 OF THREE METHODS OF DETERMINING DEOXYRIBONUCLEIC ACID 547 Manson and Lampen15 tested both deoxyribonucleosides and deoxyribonucleotides by this method.They found that the deoxyribosides of guanine and hypoxanthine and thymidine gave the same amount of colour per micromole of S-deoxy-r~-ribose, but that with cytosine deoxyriboside the colour developed more slowly and was considerably less. They also found that adenine and guanine deoxyribonucleotides gave approximately the same colour intensity as an equimolar amount of deoxyribose, the thymine derivative gave 74 per cent. of the theoretical value and cytosine deoxyribonucleotide gave no colour. For some extracts the difference was not great, but with others the colour was orange instead of purple-pink. This was partly because some extracts gave a pale orange-brown colour on addition of 70 per cent.sulyhuric acid with no cysteine present. Extracts from cabbage leaf fibre gave a bright yellow at once on addition of sulphuric acid. The rate of colour development in a fibre extract was similar to that of DNA. Fig. 4 shows the absorption curves of the colours given Most fibre extracts gave a colour different from that with DNA. 400 500 653 700 Wavelength, mp Fig. 4. Comparison of absorption curves of colours given by deoxyribose, DNA and a fibre extract with cysteine hydrochloride and sulphuric acid. Curve A, 1 mg of DNA; curve B, 240 pg of deoxyribose; curve C, fibre extract - 1 I I 2 4 6 8 1 Amount of tryptophan, mg 1 Fig.5. Dependence of colour intensity given by DNA on the amount of tryptophan used in the Cohen method. 500-pg lots of DNA, 1 ml of 60 per cent. perchloric acid, 1 per cent. tryptophan solution and water to give 3 ml. Heated 20 minutes in bath of boiling water, cooled and then diluted with 6 ml of 20 per cent. perchloric acid by a fibre extract, deoxyribose and DNA with cysteine and sulphuric acid. Stumpf’s observation that the absorption maximum is at 490 mp was confirmed, but there was also a shoulder at 560mp, which was not shown on his curve. This shoulder was also found in the deoxyribose curve. After several hours the colour of the solution became more purple and the shoulder at 560 mp became more pronounced. The fibre extract had a maximum at 490 mp, but there was also considerable absorption in the 400 to 450-mp region.INTERFERING SUBSTANCES- Five milligrams of yeast RNA, starch, glucose, arabinose, pectin and galacturonic acid gave no colour with cysteine and sulphuric acid. Digitoxose, 0.5 mg, gave a strong brown colour. Agar gave no colour on standing for 30 minutes and a yellow-green after 18 hours. 3 : 6-Anhydro-a-methyl-~-galactoside gave a yellow after 30 minutes and a green after 18 hours. The hepta-acetyl derivatives of glucose, mannose and galactose and the hexa-acetyl derivatives of rhamnose and arabinose gave no colour in 30 minutes. On standing for 18 hours the hexa-acetyl rhamnose was salmon-pink, but the other sugars were faintly straw coloured.548 HOLDEN: A COMPARISON OF THE APPLICABILITY TO PLANT EXTKACTS [VOl.78 TRYPTOPHAN - PERCHLORIC ACID METHOD To 1 ml of solution containing DNA were added 0.2 ml of 1 per cent. tryptophan in 0.01 N sodium hydroxide solution and 1.2 ml of 60 per cent. perchloric acid, the mixture was heated at 100" C for 10 minutes and then rapidly cooled. To eliminate colours produced by substances other than DNA the pink colour caused by DNA was extracted with isoamyl alcohol, and the solution was used for absorptiometric determination. A 1 per cent. w/v solution of DL-tryptophan could not be made at room temperature with 0.01 N sodium hydroxide, so a 0.1 N solution was used. With a final concentration of 30 per cent. of yerchloric acid in the reaction mixture, DNA gave an orange colour of low intensity and the reagent blank was coloured.A final concentration of 20 per cent. gave the most satisfactory results, as the blank was colourless and the colour given by DNA was more intense and a clear pink. The intensity of colour was also affected by the amount of tryptophan and the time of heating. Fig. 5 shows the effect on the intensity of colour developed by 500pg of DNA with 1 to 10-mg amounts of tryptophan. Fig. 6 shows the effect of time of heating in a bath of boiling water on the development of colour with DKA, deoxyribose, hypoxanthine The method as described by Cohen6 was as follows. Some modifications were found necessary. Time of heating, minutes Fig. 6. Effect of time of heating with tryptophan and perchloric acid on the develop- Curve A, deoxyribose; curve B, hypoxanthine deoxyriboside; curve C, DNA; curve D, 72 pg of deoxyribose and amounts of other substances calculated to con- ment of colour by DNA, deoxyribose, thymidine and hypoxanthine deoxyriboside.thymidine. tain 72 pg of deoxyribose deoxyriboside and thymidine. Cohen6 found that when DNA was heated with 30 per cent. perchloric acid for 40 minutes at 100" C all of the deoxyribose present reacted. With the conditions used in this investigation, hypoxanthine deoxyriboside gave over 90 per cent. of the theoretical amount of deoxyribose and thymidine only about 55 per cent. An increase in the concentration of perchloric acid to 30 per cent. did not give an increase of the optical density, but caused a brown coloration. There was little further increase in optical density when heating with tryptophan and 20 per cent.perchloric acid was continued for longer than 20 minutes. One millilitre of solution (or 1 ml of water for the blank), 1 ml of 1 per cent. w/v tryptophan solution in 0.1 N sodium hydroxide solution and 1 ml of 60 per cent. perchloric acid were mixed and heated in glass-stoppered tubes in a bath of boiling water for 20 minutes. After they had been cooled in ice-water the samples were diluted with 6 ml of 20 per cent. perchloric acid and the optical density read with an Ilford No. 623 filter in the absorptiometer. The extraction with isoamyl alcohol was omitted. There was a linear relationship between amount of DNA and optical density with 0-05 to 0-5 mg of DNA. The standard conditions of the method used in this investigation were as follows.Sept., 19531 OF THREE METHODS OF DETERMINING DEOXYRIUONUCLEIC ACID 549 The colour given by fibre extracts was often indistinguishable from that of a DNA standard, but was sometimes a brownish-pink.A comparison of the absorption curves of the colours given by a fibre extract, DNA and deoxyribose in this method is shown in Fig. 7. The absorption maximum for DNA and deoxyribose was at 500mp as found by Cohen.6 The fibre extract also had an absorption maximum at 500 mp, but as in the other methods there was strong absorption in the 400 to 450-mp region. The rate of colour development in a fibre extract was similar to that of DNA on heating for as much as 20 minutes, but when 1 I I I 400 500 600 700 Wavelength, mp Fig.7. Comparison of absorption curves of colours given by deoxyribose, DNA, Curve A, fibre extract; curve B, 1.12 rng of DNA; curve C, 150 pg of deoxyribose: galacturonic acid and a fibre extract with tryptophan and perchloric acid. curve D, galacturonic acid heating was continued for longer the fibre extract developed a greater colour intensity than the corresponding DNA standard and the difference in hues increased. INTERFERING SUBSTANCES- The following absorptiometer readings were obtained when 5 mg of various substances were tested under the standard conditions of the method; they are compared with the reading for 200pg of DNA. Absorptiometer reading-DNA, 14.0; yeast RNA, 3.9; pectin, 3.2; starch, 5 - 5 ; glucose, 4.2; arabinose, 3.7; galacturonic acid, 9.8. AS in the diphenylamine method, galacturonic acid interferes more than the other sub- stances tested.The absorption curve for galacturonic acid is shown in Fig. 7. Digitoxose gave a pink colour similar to that given by DNA, but with an absorption maximum at 510 mp instead of 500 mp. Two milligrams of agar and 0.56 mg of 3 :6-anhydro-a-methyl-~- galactoside gave dark brown colours with identical absorption curves. Hexa-acetyl rhamnose gave a light brown colour, but the hexa-acetyl derivative of arabinose and the hepta-acetyl derivatives of glucose, galactose and mannose gave only faint colours. COMPARISON OF THE RESULTS BY THE THREE METHODS The DNA contents of some fibre extracts determined by the three methods are compared in Table I. This shows that there is reasonable agreement between the results obtained with the diphenylamine and tryptophan methods, but that with the cysteine method the results are invariably lower and not a constant proportion of those found by the other methods.The difference between the measured phosphorus content of the solution and the value calculated for DNA phosphorus from the absorptiometric DNA determinations depends on how thoroughly the RNA has been removed from the tissue before extraction of DNA. In some experiments, when the preliminary removal of RNA was not thorough, the agreement between the diphenylamine and tryptophan methods was not as good as that shown in550 HOLDEN p o l . $8 Table I. The tryptophan method gave higher results, which suggested that the extraction of RNA is accompanied by removal of other interfering substances. In a perchloric acid extract of leaf fibre the DNA will be present mainly as nucleotides.In the diphenylamine method the purine-bound deoxyribose only is estimated, and in the other two methods the deoxyribose attached to the purines reacts completely and the pyrimidine-bound only partly. If the ratio between purine and pyrimidine nucleotides is constant for all extracts and also the same in the DNA used for standards no difficulty arises. However, differences in the ratio may occur, and this should be borne in mind particularly in estimations of DNA in extracts prepared in different ways. The interference of galacturonic acid in the diplienylamine and tryptophan methods might be considerable in extracts from plant tissues if treatments, such as heating at a high temperature in concentrated acid, which cause pectin breakdown are used.But it is possible with caution to use both methods with most extracts and get fair agreement between them. The disadvantages of the cysteine - sulphuric acid method are that addition of 70 per cent. sulphuric acid alone produces colours in many extracts and the results are erratic and bear little relation to those obtained by the other methods. TABLE I COMPARISON OF RESULTS ON PERCHLORIC ACID EXTRACTS O F TOBACCO LEAF FIBRE BY THREE METHODS OF DNA DETERMINATION Phosphorus in DNA Phosphorus measured, 17 21 24 32 36 44 49 60 71 82 tLg Per ml By diphenylamine By tryptophan method, method, 15 18 15 16 22 25 30 34 34 33 39 40 44 43 56 57 66 64 75 80 Pg Per ml tLg Per ml By cysteine method, 9 11 20 21 31 27 25 40 45 58 PLg Per ml REFERENCES 1. Dische, Z., Mikrochemie, 1930, 8, 4. 2. 3. 4. 5. 6. Cohen, S. S., ]bid., 1944, 156, 191. 7. 8. 9. 10. 11. 12. 13. 14. 15. Ogur, M., and Rosen, G., Arch. Biochem., 1950, 25, 262. McClendon, J . H., Amev. J . Bot., 1952, 39, 276. Dische, Z., Proc. SOC. Exp. Biol. Med., 1944, 55, 217. Stumpf, P. K., ,J. Biol. Chem., 1947, 169, 367. Holden, M., Biochem. J., 1952, 51, 433. Pirie, N. W., Brit. J . Exp. Path., 1936, 17, 269. Hanes, C. S., Biochem. J., 1929, 23, 99. Tracey, M. Ti., Ibid., 1948, 43, 185. Weeks, L. F., and Friminger, 13.. I., I n d . Eng. Chem., Anal. Ed., 1942, 14, 760. Deriaz, R. E., Stacey, M., Teece, E. G., and Wiggins, L. F., J . Chem. Soc., 1949, 1222. Davidson, J. N., “Biochemistry of the Nucleic Acids,” Methuen & Co. Ltd., London, 1950. Overend, W. G., J . Chem. SOC., 1951, 1484. Manson, L. A., and Lampen, J. O., J . Biol. Chew., 1951, 191, 87. ROTHAMSTED EXPERIMENTAL STATION HARPENDEN, HERTS. January 12th, 1953

ISSN:0003-2654    

DOI:10.1039/AN9537800542

出版商:RSC    

年代:1953

数据来源: RSC