Determination of glycoalkaloids in potatoes - Springer Link

13 downloads 3943 Views 347KB Size Report
HaPO4 and 1% paraformaldehyde could produce a color reaction with .... 27.4. 20.0. 79 (Ave.) L3. 27.4. 24.0. H1. 85.0. 62.0. H2. 82.0. 61.0. 74 (Ave.) H3. 84.0.
302

[Vol.

AMERICAN POTATO JOURNAL

DETERMINATION

OF

GLYCOALKALOIDS

IN

49

POTATOES

(S. TUBEROSUM) W I T H A B I S O L V E N T E X T R A C T I O N METHOD S. L. W A N G 2, C. L. BEDFORD a, AND N. R. TI-IOMPSON 4

ABSTRACT A solvent mixture of methanol and chloroform (2:1) was employed in the determination of glycoalkatoid in potatoes. The result obtained from bisolvent extraction was higher than that obtained from Soxhlet extraction with 3% acetic acid in ethanol. A more reproducible result was obtained with the steam-blanched surface sample than with the diced raw sample. Solanine and chaconine, the two components of glycoalkaloid in potato, could only be differentiated with thin layer chromatography, and they were determined colorimetrically. B5141-6 potatoes had much higher glycoalkaloid than did Russet Burbank and Kennebec potatoes as determined by the bisolvent extraction method using sliced samples. INTRODUCTION A number of methods have been employed in the determination of

"Solanine" in potatoes. Earlier methods such as direct weighing after precipitation (Morgenstern 9) and indirect estimation by measuring the reducing power of sugars after hydrolysis (Conner 5) are no longer used. Colorimetric methods are commonly used although a polarographic method is also available. (Pierzchalski and Mrozowska 12). Alberti (1937) reported that a mixture of conc. H,,SO4 and formaldehyde, the Marquis reagent, could be employed for the colorimetric determination of solanine. This method was modified by Pfankuch (11), Rooke et al. (13), Wolf and Duggar (19), Dabbs and Hilton (6) and Baker et al. (3). Clarke (4) discovered that a mixture of 85% HaPO4 and 1% paraformaldehyde could produce a color reaction with solanine and chaconine. Wierzchalski and Wierzchowska (18) used a mixture of SbCla and conc. HCI for the determination of solanine. Most of the recent quantitative analyses for total solanine have been done on the basis of the color reaction between conc. H3PO4 and paraformaIdehyde because of its simplicity and safety (Patt and Winkler, 10; Schwarze 15; Vecher et al. 16). Potato samples in most cases were extracted with various concentrations of acetic acid in either ethanol or 1-I20, A Soxhlet extraction procedure was reported to recover 100% of the added pure solanine, and has been widely used. (Baker et al. 3; Part anl Winkler 10). Sachse and Bachmann (14) reported recently that refluxing potato samples repeatedly with 80% ethanol yielded more glycoalkaloid than did 1Journal article number 5934 of the Michigan Agricultural Experiment Station. Accepted for publication February I1, 1972. ~Partial fulfillment of Ph.D. requirement, Department of Food Science and Human Nutrition. aProiessor, Department of Food Science and Human Nutrition. 4professor, Department of Crop and Soil Sciences, Michigan State University, East Lansing, Michigan 48823.

WANG, et al: GLYCOALKALOIDSIN POTATOES

1972]

303

the Soxhlet extraction. All of these extraction methods were carried out using a polar solvent system. In this paper a bisolvent (polar and non-polar) extraction procedure is described for the determination of glycoalkaloid content of potatoes. MATERIALS AND METHODS

I.

Samples:

Russet Burbank, Kennebec and B5141-6 potatoes were used. Surface samples were prepared by scraping the 2 to 3 mm thick surface from the entire tuber of the steam-blanched and hand-peeled potatoes. Diced samples were taken from peeled and trimmed raw potatoes. Sliced samples were slices cur cross-sectionally between the bud and the stem end of the potato tuber. Russet Burbank potatoes were treated with white light for 6 days to obtain a high level of glycoalkaloid in the surface of the tuber. Samples with high and low levels of glycoalkaloid were obtained from the surface of the potatoes with and without exposure to white light, respectively, II.

Extraction:

Five g samples were weighed in triplicate and blended with 100 ml of a mixture of methanol-chloroform (2:1) in a semi-micro stainless steel blender assembly for 4 to 5 rain. The suspension was filtered through a sintered glass Buehner funnel (medium porosity), and the residue rinsed twice with a 25 ml portion of the same solvent mixture. The filtrate was transferred into a 500 ml separatory funnel, 60 ml of 0.8% Na2SO4 added, and the mixture vigorously shaken. The layers were allowed to separate and the lower layer (chloroform) was discarded. The methanolic layer was re-extracted with 10 ml more of 0.8% Na,2SO4, and the lower layer discarded again. The methanolic solution was collected into a 250 ml beaker, and the separatory funnel rinsed with 1% acetic acid. This solution was concentrated on a steam bath to near dryness and made to 10 or 25 mt depending on the concentration of glycoalkaloid in the samples. Some samples were extracted with 3 % acetic acid in ethanol for 18 hours with the Soxhlet extractors, and glycoalkaloid sample solution prepared according to the method described by Baker et al. (1955). III.

Determination and identification of glycoalkaloid:

Five ml of the salnple solution was pipetted into a 40 ml centrifuge tube (tapered bottom), and the ptI adjusted to 9.4 with 5M N H , Ott. After the tube was heated at 80 C in a water hath, glycoalkaloid was flocculated and centrifuged. The precipitate was washed once with a small amount of water. To the precipitate, 5 ml of 8 5 ~ HaPO4, followed immediately by 5 drops of saturated paraformaldehyde solution was added. The amount of glycoalkaloid was determined colorimetrically at 600 nm with a Beckman DB spectrophotometer (Wang 1970). The glycoalkaloid extracted with the bisolvent or the Soxhlet extraction method was chromatographed on a Silica gel plate using a solvent sys-

304

AMERICAN POTATO JOURNAL

[Vol. 49

tern, methanol-ethyl acetate-acetic acid-water (20:30:10:1) ( V / V ) . The separated glycoalkaloids were then treated with the following reagents, and their absorption spectra recorded with a Beckman DB Spectrophotometer. 1. Marquis reagent: 6 ml of cone. H2SO4 plus 3 ml of 1% of HCHO. 2. Clarke reagent : 5 ml of cone. HaPO4 plus five drops of saturated paraformaldehyde solution 3. Cone. HaPO4. RESULTS AND DISCUSSION~

Using the surface samples with a relatively high level of glycoalkaloid, the extraction yield obtained with the Soxhlet method was about 75% of that with the bisolvent method (Table 1). Higher yields were also obtained by the bisolvent method for the diced samples (Table 2). However, the yield difference was not significant when the glycoalkaloid content of the sample was low. Good agreement in results was obtained between the replicates of the surface samples, but not the diced samples. The lack of agreement between the replicates of the diced samples was due to the nonhomogeneous distribution of the glycoalkaloid in the tuber (Lampitt et al. 1943). The surface area from steam-blanched, peeled potatoes, thoroughly mixed, provided very homogeneous samples. Analyses of the surface samples from various areas of the tubers indicated that the glycoalkaloid tended to be slightly variable (Table 3). Two sets of surface samples with different levels of glycoalkaloid were used in studying the percent recovery of solanine by the bisolvent extraction method. One mg of pure a-solanine, equivalent to 73% of the glycoalkaloid in the low glycoalkaloid samples and 24% in the high glycoalkaloid samples, was completely recovered (Table 4). The results showed that the bisolvent extraction with methanolchloroform (2:1) was more efficient than the Soxhlet extraction with the polar solvent system. In samples low in total glycoalkaloids the Soxhlet extraction had given slightly higher readings. Since the bisolvent extraction provided a purer glycoalkaloid precipitate, the difference in color appearance was obvious (Table 2). Several advantages were found with the bisolvent extraction method. By blending the sample with a dry solvent mixture for 4 to 5 min., the sample was dehydrated so that little starch or protein was dissolved; filtration was greatly improved. Separation of the chloroform fraction from the methanolic fraction with dilute Na2SO4 removed the lipid material, which is a source of contamination. Sachse and Bachman (14) used ethyl ether to remove the lipid material. Polymerization of pigments or browning, which took place readily in aqueous solution upon prolonged heating, was eliminated by shortening the heating time in extracting, precipitating and flocculating the glycoalkaloid. Also, use of centrifugation in collecting the precipitates minimized the loss of the fluffy glycoalkaloid flocculates. In the thin layer chromatographic study, both a-solanine and a-chaconine were found in all samples. No attempt was made to establisl~ a quantitative relationship between these spots. Absorption spectra of a-solanine were identical to those of a-chaconine with all three reagents tested (Fig. 1). With the Clarke reagent both

1972]

WANG,

TABLE

et al: GLYCOALKALOIDS IN POTATOES

305

1.--GIycoalkaloid content of potato surface samples extracted ~vith the bisolvent and the Soxhlet extraction method.

Sample

Bisolvent method wet basis mg/lO0 g

Soxhlet method wet basis mg/lO0 g

Yield (Soxhlet) Yield(Bisolvent) %

L1

27.4

21.0 20.0 24.0

79 (Ave.)

L3

27.4 27.4

H1 H2 H3

85.0 82.0 84.0

62.0 61.0 63.0

74 (Ave.)

L2

Note: L and I-I designate potato surface samples with low and high level of glycoalkaloid respectively. TABLE

2.--GIycoalkaloid content of diced potato samples determined ~ith the bisolvent and Soxhlet extraction methods.

Sample

Bisolvent method wet basis mg/100 g

L-45 L-45 L-45

6.8 7.8 9.2

6.2 6.0 6.2

L-65 L-65 L-65

16.0 20.8 24.8

7.4 11.6 7.8

RB-40 RB-40 RB-40

Soxhlet method wet basis rag/100 g

Trace Trace 0.6

1.4 2.0 2.4

Note: "L-45 and L-65 are B5141-6 potatoes stored at 45F and 65F respectively, and RB-40 designates Russet Burbank potatoes stored at 40F. TABLE

3.--GlycodkaIoid content of ~urface samples from various areas of the potato. Potato variety

Stem end

Bud end

Other area

25.0 37.0 70.0

23.0 45.0 60.0

Glycoalkaloid content mg/100 g wet basis Russet Burbank Kennebec B5141-6

28.0 62.0 64.0

306

AMERICAN POTATO JOURNAL

[Vol. 49

TABLE 4.--Recovery of 1 m 9 added solanine using the bisolvent extraction

method (surface sample used as solanine carrier).

Sample

Glycolalkaloid With addition 1 nag solanine Wet basis Wet basis rag/5 g mg/5 g

Recovery mean %

L1 L2 L3

1.37 1.37 1.37

2.45 2.40 2.40

105

H1 H2 H3

4.25 4.10 4.20

5.10 5.30 5.30

105

Note: L and H designate potato samples with low and high level of glyeoalkaloid respectively. Chaconine

/,,-',,\

/~ I

L.

.o

I,t

.

700 Figure I.

600

,500

,

400

Wavelength nm Absorption spectra of solanine and chaconine after reacting with (a) the Clarke reagent, (b) the Marquis reagent, (c} 85~, H3PO 4.

glycoalkaloids showed the maxinmm absorptions at 660, 600 and 512 nm. Use of 600 nm for a quantitative determination of the glycoalkaloids was appropriate. This proved that earlier solanine data, determined colorimetrically included both solanine and chaconine. The glycoalkaloid analyses on the slices of the three potato varieties showed that the average glycoalkaloid level of B5141-6 was more than double that found in either Russet Burbank or Kennebec (Table 5). The high level of glycoalkaloid in B-5141-6 was in agreement with the results reported by Zitnak and Johnston (20). The distribution studies showed that the glycoalkaloid was concentrated in the cortical region of Russet

1972]

WANG,

et al: GLYCOALKALOIDSIN POTATOES

307

TABLE 5.--Glycoalkaloid contents of Russet Burbank,

Kennebec and B5141-6 potato slices determined by the bisolvent extraction method) Glycoalkaloid Content Potato variety

Wet basis Russet Burbank 5.7 8.9 5.9 7.5 6.7 Mean 6.9 Kennebec

Mean B5141-6

Mean

Cortical region 2 mg/100 g

Whole slice

Pith

Dry basis 24.9 45.2 35.5 36.2 30.3 .34.4

Wet basis 7.3 13.1

Dry basis 31.8 66.4

Wet basis ,Trace Trace

Dry basis Trace Trace

8.4 7.3 5.5 8.5

8.3 14.9

46.9 76.8

Trace Trace

Trace Trace

7.7

47.5 37.6 29.7 45.2 42.4

28.3 28.4 18.8 24.0 22.2 24.3

127.5 122.4 80.3 96.8 85.4 102.5

32.3 31.1

145.4 134.0

13.9 10.1

62.6 43.5

1Potatoes have been stored in the dark at ,45F for 50 days. 2Cortical region includes cortex, outer and inner phloem and xylem (Identical to surface sample). B u r b a n k and Kennebec, while in B5141-6 it was distributed t h r o u g h o u t the tuber. T h e s e results indicated that the analyses of slices could be used as a rapid sampling m e t h o d for screening of potato varieties for their glycoalkaloid level. LITERATURE CITED 1. Alberti, B. 1932. Zum Nachweis yon Solanin. Z. Lebensm.-Untersuch.Forsch. 64: 260. 2. Allen, E. H. and J. Kuc. 1968. ~-solanine and a-chaconine as fungitoxic compounds in extracts of Irish potato tubers. Phytopathology 58(6) : 776. 3. Baker, L. C., L. H. Lampitt and C. B. Meredith. 1955. Solanine, glycoside of the potato. III. An improved method of extraction and determination. J. Sci. Food Agri. 6: 197. 4. Clarke ,E. G. C. 1958. Identification of solanine. Nature 181: 1152. 5. Conner, H . W . 1937. Effect of light on solanine synthesis in the potato tuber. Plant Physiol. 12: 79. 6. Dabbs, D. H. and R. J. Hilton. 1953. Method of analysis for solanine in tubers of Solanum tuberosum. J. Technol. 31: 213. 7. Lampitt, L. H., J. H. Bushill, H. S. Rooke and E. M. Jackson. 1943. Solanine, glycoside of the potato. II. Its distribution in the potato plant. J. Soc. Chem. Ind. 62: 48. 8. Lowry, R. R. 1968. Ferric chloride spray detector for cholesterol and cholesterol esters on TLC. J. Lipid Res. 9: 397.

308 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

AMERICAN POTATO JOURNAL

[Vol. 49

Morgenstern, F. Von. 1907. Uber den Solaningehalt der Speise-und FutterKartoffeln and uber den Einfluss der Bodenkultur auf die Bildung yon Solanin der Kartoffelpflanze. Land-wirschaftlichen Versuchs-Station 65: 301. Patt, P. and W. Winkler. 1960. Zur Darstellung und Bestimmung yon Solanin dureh Ionenaustausch. Arch. Pharm. 293: 846. Pfankuch, E. 1937. Die photometrische Bestimmung von Solanin. Biochem. Z. 295: 44. Pierzchalski, T. and A. Mrozowska. 1%8. Polarographic determination of glyeoalkaloids. Chem. Anal. ( W a r s a w ) 1 3 ( 2 ) : 367. Rooke, H. S., J. H. Bushill, L. H. Lampitt and E. M. Jackson. 1943. Solanine, glycoside of the potato. I. Its isolation and determination. J. Soc. Chem. Ind. 62 : 20. Sachse, J. and F. Bachmann. 1969. Uber die Alkaloidbestimmung in Solanum tuberosum L. Z. Lebensm.-Unters, und-Forchung. 141: 262. Schwarze, P. 1962. Methoden zum Solinnachweiss und zur Solaninbestimmung in Kartoffelzucht Material. Der Zfichter 32: 155. Veeher, A. S. and O. K. Vasil kevich. 1%7. Solanine content in the most important and new potato varieties in the BSSR. Beloruss S. S. R. 1 1 ( 4 ) : 362. Wang, S. L. 1970. Glycoalkaloid and demethyl sterol content in some potato varieties and clones. Ph.D. Thesis, Michigan State University. Wierschowski, P. and Z. Wierzchowska. 1%1. Colorimetric determination of solanidine and solanine with SbCh. Chem. Anal. 6: 579. Wolf, M. J. and B. D. Duggar. 1946. Estimation and physiological role of solanine in the potato. J. Agri. Res. 73: 1. Zitnak, A. and G. R. Johnston. 1970. Glycoalkaloid content of B5141-6 potatoes. Amer. Potato J. 47: 256.