J. Inst. Brew., July-August, 1980, Vol. 86,/y?. 171-173 171 DETERMINATION OF DNA AND OF OTHER NUCLEIC ACID FRACTIONS IN SACCHAROMYCES CEREVISIAE By William E. Trevelyan {Tropical Products Institute, 56-62 Gray's Inn Road, London WC1X8LU) Received 24 October 1979 After extraction overnight of Saccharomyces cerevisiae (baker's yeast) with dilute alkali, to remove nucleotide pools and RNA, acidification gave an insoluble residue from which DNA guanine and adenine were extracted, in the free form, by 1M perchloric acid at room temperature, and deter mined by cation-exchange chromatography. Percentage DNA in the yeast was computed as 0062 x pmol/g of guanine + adenine. Preliminary extraction of nucleotides was unnecessary, and gave low results if acid reagents (perchloric or trichloroacetic acids) were used. Several reagents were tested for their ability to extract the nucleotide pool from intact yeast without significant effect on cell RNA content. The content of guanine + adenine in pool-free yeast was taken as giving the sum of RNA + DNA purines, RNA purines being found by difference. Key words: yeast, analysis, DNA, nucleic acids, purines. Introduction The nucleic acids of yeast1- have technological as well as scientific importance. Single-cell protein, of which the brewing industry is one of the largest producers,13 has to be processed to reduce its high content of nucleic acid (RNA) if it is to be suitable for human consumption,8 though the same problem does not arise with animal feeds.14 Yeast is also a source of ribonucleic acid used in the production of flavour nucleo tides.7 DNA, which occurs in yeast in relatively small amount, is not produced in bulk; but its manipulation, by the tech niques of genetic engineering, is now attracting considerable attention,20 while the amount per cell, and the base composi tion (% GQ, of yeast DNA have important implications, e.g., in taxonomy." The DNA content of yeast is usually determined, after its extraction from yeast with hot, dilute perchloric acid,10 by what Carter1 describes as '. . . the laborious and somewhat inaccurate diphenylamine method'. An alternative method was suggested by the rinding that treatment of yeast with Im perchloric acid (PCA) for 18 h at room temperature de- purinated its DNA, releasing free adenine and guaninc, in the molar ratio of 1-7:1, characteristic of yeast deoxyribonucleic acid.10 The DNA content, as estimated from the amount of free adenine + guanine extracted by PCA, agreed with that given by the standard diphenylamine procedure in the case of yeast preparations from which nearly all the RNA had been removed by autolysis. Unfortunately, RNA is depurinated to some extent by IM PCA, so that the method failed when it was applied to the original, non-autolysed yeast, which contained about twenty times as much RNA as DNA. It has now been found that RNA can be efficiently removed from the cells of Saccharomyces cerevisiae by dilute sodium hydroxide, and remains in solution when DNA (largely, but not entirely, bound to the cells) is precipitated by acid (Schmidt-Thannhauser procedure). DNA can then be readily determined after extraction of the insoluble residue with 1m perchloric acid. RNA, the major nucleic acid fraction of yeast, can also be determined by purinc analysis. The nucleotide pool is first re moved from a duplicate sample of yeast by any of a number of possible methods, when 1m perchloric acid at room tempera ture extracts from the pool-free cells both the purines of DNA, and partially degraded RNA in the form of an ill-defined mixture of oligonucleotidcs. Hydrolysis, by heating the extract for 1 h at 100°C, liberates all purines in the free form, allowing the determination of the sum of the purines of DNA and RNA. The content of RNA purines, and hence the amount of RNA in the yeast, is then obtained by difference. All the methods for extracting the nucleotide pool which were tried failed to effect an absolute separation of nucleo tides from nucleic acids, though this did not cause any serious error in the determination of RNA content. Methods Purine estimation.—A 2-5 ml sample loop, made by connect ing two 3-way valves (Pharmacia LV-3) with the appropriate length of 1 -S mm bore PTFE tubing, was charged with the sample. One end of the loop was then connected to a peristaltic pump, which delivered 40 ml/h of a filtered, de-aerated solu tion of K.H2POi(0-5M,pH 4-3), and the other to a 110 x 10mm column of Bio-Rad AG SOW X4 cation-exchange resin, main tained at 32°C. Pyrimidine nuclcotides were eluted early on, and followed by guanine and then adenine in well-defined peaks, which were recorded by an ISCO Model UA5 ultra violet monitor (254 nm). After 3 h, the column was ready to receive the next sample. Over a period of several days, peak heights were reproducible to within 1 mm, corresponding to about 0-1 ixmol purinc/g yeast dry matter (DM) in the case of DNA analysis. Comparison was made with the peak heights given by a standard solution of guanine + adenine,17 0-5 /xmol/ml in Im PCA, which was mixed with an equal volume of Im K2HPO4 and, after the precipitate of potassium per- chlorate had been removed, diluted appropriately with 0-5m KH2PO4. Yeast.—Eighteen g of commercial pressed baker's yeast (Distillers Company Ltd) was washed three times in succession by centrifuging a suspension in 60 ml water, and made to 100 ml. Five ml of suspension, delivered by pipetting syringe, contained approximately 250 mg of dry matter (residue after drying for 20 h at IO5°C); duplicate determinations differed by 0-3 mg (mean, N = 9). Total purine content.—Twenty ml of 1-25m PCA were added to S ml of yeast suspension contained in a screw-capped (universal) bottle; this was clipped to a motor-driven shaft and rotated end-over-end for 18 h. The yeast cells were sedimented by centrifuging, and 10 ml of the supernatant heated for 1 h at 100°C (water-bath, screw-capped tube) to obtain purines in the free form. Before the solution was applied to the column, perchlorate was precipitated with K2HPO4, as above. Neglect of the volume occupied by the yeast solids intro duced an error of less than 1 %. An alternative procedure, convenient when the yeast sample was in the form of a centrifuged pellet, was to extract twice with Im PCA (10 ml for 18 h, then 10 ml for 0-5 h; combined supernatants + 2-5 ml of 2m PCA diluted to 25 ml). Extraction of nucleotide pool.—Several methods were tried, of which the following gave reproducible results. (A). Five ml of yeast suspension were delivered into a thin-walled tube, and stirred by means of a 2 mm glass rod attached by a flexible coupling to the shaft of an Electrothermal Mini- stirrer. The tube was immersed for 5 min in a bath of water at 80°C, and then cooled in ice. The stirrer was washed off with 5 ml of water; then the yeast was recovered by centrifuging, and re-extracted for 10 min with 10 ml of cold water. (B). The sample of yeast suspension was mixed end-over-end for 15 min with 5 ml of trichloroacetatc buffer (1 mol TCA + 0-5 mol NaOH/litrc), centrifuged, and the cells re-extracted 172 trevelyan : determination of dna in Saccharomyces cerevisiae [J. Inst. Brew. with 10 ml of water (10 min). (C). The sample was extracted for IS min with 5 ml of a solution containing 0-4 mol of perchloric acid -(- 4-0 mol of acetic acid/litre, and re-extracted with 10 ml of 0-2m PCA. (D). The sample was centrifuged and the supernatant discarded. With a glass rod, the yeast pellet was stirred for 5 min with 1 ml of 2m PCA; the rod was washed off with 8 ml of water, the suspension mixed end-over- end for IS min, the yeast centrifuged down and re-extracted with 10ml of 02m PCA. To determine total guanine + adenine in the extracts, they were made 1m with respect to PCA, and hydrolysis carried out at 100°C, as above. When TCA was the cxtractant, this had to be done in an open tube, to allow the escape of chloroform resulting from the decomposition of the acid. DNA + RNA purine content.—With the aid of a plastic Pasteur pipette, the pellet of pool-free yeast cells was trans ferred (in water) to a 25 ml volumetric flask. Five ml of 5m PCA were added, followed by water to the mark. The suspen sion was poured into a screw-capped bottle and then treated as under total purine content. Alternatively, the successive extrac tion procedure was used. DNA purine content.—Five ml of yeast suspension (or 250 mg of freeze-dried yeast) were measured into a IS ml polypropylene centrifuge tube; 2-5 ml of 1m NaOH were added, and water to 10 ml. The tube was stoppered, and the contents agitated gently overnight (end-over-end mixer). Perchloric acid (2-5 ml of a 1-5m solution) was added, and the suspension mixed for 10 min. The insoluble residue was centrifuged down, washed twice with 10 ml of 01m PCA, and extracted overnight with 8 ml of 1m PCA. The supernatant was decanted into a 25 ml graduated cylinder, and followed by a second extract (4 ml Im PCA -t- 4 ml water, 30 min); the combined extracts were made to 17 ml. Then 8 ml of 1-5m K2HPO4 were added, KC1O4 removed by centrifuging, and free purines determined on the supernatant. A shorter pro cedure is to precipitate the acid suspension of yeast directly with Kj,HPO4 without first centrifuging off the cells; but once the extracts are neutralized, the analysis should be completed the same day. Results DNA estimation.—The method was tested on four separate batches of pressed yeast, all of which had a total guanine -|- adenine content in the region of lOOjamol/g dry matter (Table II). Between analyses, the yeast was stored at 4°C (for periods of up to 3 weeks). Free guanine in the DNA fraction amounted to 1 -8—2-1 /imol/g yeast DM, and adenine, 3-O-3-5 /xmol/g. For 12 estimations on batch Y91115, the standard deviation was ±005 and ±007/xmol/grespectively. The mean molar ratio adenine/guanine was 1-73 ± 0-02 (SE, 12 determinations) corresponding to a GC content of 37%, in agreement with published data for yeast DNA.10 Theoretically, a double- stranded DNA of this base composition should contain 1 mole purine in 618 g; hence the percentage by weight of DNA in the samples of yeast examined = 00618 x ^mol/g (guaninc + adenine) = 0-30-0-35 %, as found previously by the diphenylamine procedure.10 No difference was found between pressed yeast and freeze-dried preparations (Table I). Total purine content of the DNA fraction, as determined after acid hydrolysis, was the same as the free purine content (Table 1). Thus digestion with alkali was effective in extracting RNA from the yeast cells and rendering it acid-soluble. If, after this step, the suspension was not acidified, the cells being simply centrifuged and washed with water, 75 % of the DNA was recovered bound to the cells (Y90621, Table I). Cold, dilute alkali is thus a poor extractant for yeast DNA, pre sumably because the molecules arc too large to pass through the pores of the cell wall.4-10 Extraction of the nucleotidc pool by water at 80°C before the alkali treatment did not affect the DNA content (Y90727, TABLE I. DNA Purinc Content of Commercial Baker's Yeast. Yeast batch Y90621 Y90727 Y90823 Y9U15 Procedure standard alkali digest not acidified yeast frcczc-dricd standard nuclcotidcs first extracted with water at 80° nuclcotidcs first extracted with 0-5m TCA standard nucleotidcs first extracted with PCA standard standard nuclcotidcs first extracted with TCA buffer alkali digest acidified with acetic acid to pH4-5 standard yeast residues not removed before pre cipitation of KCIO4 Guaninc urnol/g Free 1-9 1-3 185 1-8 18 1-6 1-8 1-6 21 21 1-9 1-9 1-8 1-8 Total 1-8 1-3 21 1-9 2-3 Adeninc ftmol/g Free 3-3 2-5 3-2 31 305 2-8 3-2 2-7 3-4 3-5 3-2 35 30 31 Total 3-2 2-5 3-5 3-2 4-3 Table I). But the results were down by 8-12% when PCA or TCA was used, possibly due to partial depurination at low pH. It was thought that the acid precipitation and washing steps in the DNA analysis procedure might cause a similar loss; but precipitation of the alkali digest with acetic acid at pH 4-5, followed by washing with 0-1m acetate buffer of the same pH, did not significantly affect the free purine figure, though total purines went up by 20%, indicating some adsorption on the cell residue of RNA breakdown products. The nucleotidc pool.—Commercial baker's yeast was found to contain 8-9 /xmol/g more adenine than guanine, a reflection of the predominance of adenine-based constituents in the nucleotide pool.0-11 (The DNA fraction contributes 1 /imol/g more adenine than guanine, but this is offset by the slight excess of guanine over adenine in yeast RNA.18) To extract nucleotides rapidly and completely from yeast is difficult,5 especially if it is required to retain DNA and RNA within the cell.8 This is due to the presence of an entire cell wall. Many reagents close up the cell structure, so that even small molecules diffuse out slowly, or may be completely trapped; ethanol is an example.10 The most commonly used extractants are the strongly acidic protein precipitants perchloric acid (PCA) and trichloroacetic acid (TCA). Their anions are chaotropic, i.e., they destabilize bio-membranes and enzyme complexes.3 As it is largely undissociated, TCA penetrates the yeast cell more rapidly. At 23°C, 0-25m TCA extracted 0-8 jttmol/g of total guanine from the washed cells of yeast Y91004 after 0-25 h. The amount extracted increased slowly with time to 1-1 /xmol/g after 0-75 h, then more rapidly, to 3-1 /xmol/g after 2 h. The amount of adenine extracted also increased, in such a way that the difference (adenine-guanine) remained constant; for 13 observations made with different concentrations of acid the mean difference was 9-6/xmoI/g (SD ± 0-3). This suggests that the increase in guanine and adeninc with increasing time was due to the extraction of RNA, in which the two purines are present in nearly equal amount. Extraction with 0-25m TCA for 0-25-O-5 h at room tem perature was considered satisfactory for removing the nucleo tidc pool from yeast preparatory to analysis for nucleic acids. Vol. 86, 1980] trevelyan: determination of dna in Saccharomyces cerevisiae TABLE II. Content of Total, Nucleic Acid, and Nucleolide Purinc in Baker's Yeast. 173 Yeast batch Y90621 Y9062I" Y90727 Y90823 Y91115 Nucleolidc pool extraction TCAB PCA W TCA TCAB PCA Total ;imolj Guaninc 46-9 470 45-1 45-6 45-2 44-9 48-2 purine 'g DM Adenine 55-2 550 521 539 53-5 54-1 56-4 RNA -I- jim Guaninc 45-3 44-5 440 44-3 47.7 b 48-1 e DNA purine ol/g DM Adenine 44-3 43-8 43-2 43-6 47.31. 47 -6C Nucleotide (*mol/g Guaninc 0-8 1-4 10 0-8 0-8 purinc DM Adenine 9-7 9-6 11-4 8-7 8-8 PCA, perchloric acid. TCA, chilled 0-5m trichloroacetic acid. TCAB, buffered TCA. W, water at 80"C. a Freeze-dried. b Standard procedure. c Sequential extraction procedure. Extraction was incomplete after 0-5 h with a weaker reagent (01 25m). A solution containing O-25m sodium trichloroacetate in addition to 025m TCA was preferred, as giving better buffering power; the pH of the extract was 11, close to the value at which yeast RNA is least soluble21 (Methods, pro cedure (B)). Perchloric acid is in many ways a more convenient reagent than TCA.8 A 0-2m solution, which established a pH of 10, was satisfactory with air-dried yeast,1" in which the cell membrane has been damaged, but was ineffective with pressed yeast, extracting only 1 /xmol of adenine/g DM after 0-5 h. A concentration of 1m PCA was required for rapid destruction of the permeability barrier in the whole population of yeast cells; by diluting the suspension to reduce PCA concentration of 0-2m after 5 min, extraction of RNA was minimized (Methods, procedure (D)). Otherwise it reached one-third of the total after 0-5 h. Another approach was to add a lipo- philic reagent (acetic acid; Methods, procedure (Q) as well as 0-2m PCA. Nucleotides diffused from the yeast cells when a suspension in water was heated rapidly to 80°C {Methods, procedure (A)). The excess of nucleotide adenine over guaninc, as found by this method, was 1-2/umol/g greater than when yeast was extracted with TCA or PCA (Table II). These methods must be regarded as giving only an approxi mate estimate of nucleotide adenine, and especially guanine, but are suitable for extracting the nucleotide pool from yeast as a preliminary to analysis of nucleic acid. The choice between them will depend on considerations such as the high UV absorption of TCA, or the interference of PCA with Kjeldahl digestion. DNA + RNA guanine and adenine.—Determinations had a standard deviation of ±0-5-1 jitnol/g. By subtracting the purines contributed by DNA, the content of RNA guanine and adenine could be found; for batch Y90727 (Table 11) this was 42-5 jumol guanine and 40-6 /nmol adenine/g DM, a molar ratio of 1-05 as compared with 1-08 found for a commercial preparation of yeast RNA." Discussion The development of the method for DNA described here now enables the three broad nucleic acid fractions usually distinguished in the routine analysis of microbial cells4 to be determined by reference to a single well-defined standard, a solution of adenine and guanine in 1m perchloric acid, which is stable over a long period of time. The separate determination of guanine and adenine has proved valuable in the interpreta tion of results. We have found the simple ion-exchange procedure for estimating these purine bases to be convenient and reliable. The numerical value of the factor 0062 used to convert DNA guaninc + adenine content (as jurnol/g yeast) to % DNA is not very sensitive to the exact base composition assumed for (double-stranded) DNA. There is a little more un certainty in the case of RNA. Data on the base composition of 5. cerevisiae RNA given by Crestfield et at.,2 correspond to the factor 0061, as do the slightly different results of Storck.15 Whatever factor is chosen, it should be regarded as a conven tion analogous to the multiplier 6-25 used to convert total N to crude protein. As nucleic acids contain about 16% N, an estimate of the protein content of yeast can be obtained by subtracting the percentage of nucleic acid from a crude protein figure calcula ted as total N in pool-free cells x 6-25. A more direct method is to extract yeast with 1m PCA (as in the estimation of total purine content), wash the cells with water, disperse them in water + a few drops of 1 m NaOH, and determine nucleic acid- free crude protein as % total (Kjeldahl) N x 6-25 (about 40% for baker's yeast). PCA at room temperature extracts little protein from yeast; in the extract from pool-free yeast, 80% of solids and of total N could be accounted for as RNA. References 1. Carter, B. L. A., Advances in Microbial Physiology, 1978, 17, 243. 2. Crestfield, A. M., Smith, K. C. & Allen, F. W., Journal of Biological Chemistry, 1955, 216, 185. 3. Hanstein, W. G., Davis, K. A. & Hatefi, Y., Archives of Biochemistry and Biophysics, 1971, 147, 534. 4. Herbert, D., Phipps, P. J. & Strange, R. E., in Methods in Microbiology, Volume 5B, edited by Norn's, J. R. & Ribbons, D. W., London, Academic Press, 1971, 308. 5. Kopperschlager, G. & Augustin, H. W., Experientia, 1967, 23, 623. 6. Kudra, R. & Edlin, G., Journal of Bacteriology, 1975, 121, 740. 7. Kuninaka, A., Kibi, M. & Sakaguchi, K., Food Technology, 1964, 18, 287. 8. Kuroiwa, Y. & Hashimoto, N., Bulletin of the Agricultural Chemical Society of Japan, I960, 24, 547. 9. Lovland, J., Harper, J. M. & Frey, A. L., Lebensmittel Wissemchaft und Technologic, 1976, 9, 131. 10. Ogur, M., Minckler, S., Lindcgrcn, G. & Lindegren, C. C, Archives of Biochemistry and Biophysics, 1952, 40, 175. 11. Oura, E. & Suomalaincn, H., Journal of the Institute of Brewing, 1967, 73, 370. 12. Piperakis, S. M. & Piperaki, F. M., Folia Biochimica et Biologica Graeca, 1977, 14, 32. 13. Pomcranz, Y., The Brewers Digest, 1976, 60, 49. 14. Roth, F. X., Kraftfutter, 1979, 62, 122. 15. Storck, R., Molecular Mycology, in Kwapinski, J. B. G. et al., Molecular Microbiology, New York, John Wiley & Sons, 1974, 423. 16. Trevelyan, W. E., Journal of the Institute of Brewing, 1966, 52, 184. 17. Trcvelyan, W. E., Journal of the Science of Food and Agricui~ lure, 1975, 26, 1673. 18. Trevelyan, W. E., Journal of the Science of Food and Agricul- lure, 1977, 28, 579. 19. Trevelyan, W. E., Journal of the Science of Food and Agricul ture, 1978, 29, 903. 20. Tubb, R. S., Journal of the Institute of Brewing, 1979,85,286. 21. Zimmermann, S. B., Biochemistry, 1973, 12, 2916. . J. Inst. Brew., July-August, 1980, Vol. 86,/y?. 171-173 171 DETERMINATION OF DNA AND OF OTHER NUCLEIC ACID FRACTIONS IN SACCHAROMYCES CEREVISIAE By William E. Trevelyan {Tropical Products Institute, 56-62 Gray's. time to 1-1 /xmol/g after 0-75 h, then more rapidly, to 3-1 /xmol/g after 2 h. The amount of adenine extracted also increased, in such a way that the difference (adenine-guanine) remained constant; for 13 observations made with different concentrations of acid the mean difference was 9-6/xmoI/g (SD ± 0-3). This suggests that the increase in guanine and adeninc with increasing time was due to the extraction of RNA, in which the two purines are present in nearly equal amount. Extraction with 0-25m TCA for 0-25-O-5 h at room tem perature was considered satisfactory for removing the nucleo tidc pool from yeast preparatory to analysis for nucleic acids. Vol. 86, 1980] trevelyan: determination of dna in Saccharomyces cerevisiae TABLE II. Content of Total, Nucleic Acid, and Nucleolide Purinc in Baker's Yeast. 173 Yeast batch Y90621 Y9062I" Y90727 Y90823 Y91115 Nucleolidc pool extraction TCAB PCA W TCA TCAB PCA Total ;imolj Guaninc 46-9 470 45-1 45-6 45-2 44-9 48-2 purine 'g DM Adenine 55-2 550 521 539 53-5 54-1 56-4 RNA -I- jim Guaninc 45-3 44-5 440 44-3 47.7 b 48-1 e DNA purine ol/g DM Adenine 44-3 43-8 43-2 43-6 47.31. 47 -6C Nucleotide (*mol/g Guaninc 0-8 1-4 10 0-8 0-8 purinc DM Adenine 9-7 9-6 11-4 8-7 8-8 PCA, perchloric acid. TCA, chilled 0-5m trichloroacetic acid. TCAB, buffered TCA. W, water at 80"C. a Freeze-dried. b Standard procedure. c Sequential extraction procedure. Extraction was incomplete after 0-5 h with a weaker reagent (01 25m). A solution containing O-25m sodium trichloroacetate in addition to 025m TCA was preferred, as giving better buffering power; the pH of the extract was 11, close to the value at which yeast RNA is least soluble21 (Methods,. 10 ml of cold water. (B). The sample of yeast suspension was mixed end-over-end for 15 min with 5 ml of trichloroacetatc buffer (1 mol TCA + 0-5 mol NaOH/litrc), centrifuged, and the cells re-extracted 172 trevelyan : determination of dna in Saccharomyces cerevisiae [J. Inst. Brew. with 10 ml of water (10 min). (C). The sample was extracted for IS min with 5 ml of a solution containing 0-4 mol of perchloric acid -(- 4-0 mol of acetic acid/ litre, and re-extracted with 10 ml of 0-2m PCA. (D). The sample was centrifuged and the supernatant discarded. With a glass rod, the yeast pellet was stirred for 5 min with 1 ml of 2m PCA; the rod was washed off with 8 ml of water, the suspension mixed end-over- end for IS min, the yeast centrifuged down and re-extracted with 10ml of 02m PCA. To determine total guanine + adenine in the extracts, they were made 1m with respect to PCA, and hydrolysis carried out at 100°C, as above. When TCA was the cxtractant, this had to be done in an