Crystalline Methyl α-Isomaltoside and its Homologs Obtained by

R. W. Jones, Allene Jeanes, C. S. Stringer, and H. M. Tsuchiya. J. Am. Chem. Soc. , 1956, 78 (11), pp 2499–2502. DOI: 10.1021/ja01592a046. Publicati...
0 downloads 0 Views 550KB Size
METHYL CPISOMALTOSIDE BY DEXTRANSUCRASE ACTION

June 5 , 1956

Experimental Isolation of C14-Labeled Cotton Seed Oil.-The seeds from ten matured cotton bolls which were treated with 33.0 MC. of ~ - g l u c o s e - 6 - C ~ (150 ~ mg.) a s described in our previous communication? were separated from the radioactive cellulose and ground in a mortar to break the hulls. The ground product (8.49 9.) was then extracted with 250 ml. of benzene in a Soxhlet extractor for three days according to the method of Koo and King.14 T h e extract was evaporated and the remaining oil was washed with a 10% solution of sodium carbonate and then with water. The dark brown product was further purified15 by adsorption on a column containing a mixture of 10 g. of Florex XXX16 and 1 g. of decolorizing carbon, and was subsequently eluted with 200 ml. of benzene. Evaporation of the solvent furnished 2.01 g. of bright yellow cotton seed oil. Conversion of the W-Labeled Cotton Seed Oil to Glycerol and Fatty Acids.-A portion of C14-labeled cotton seed oil (1.119 8.) was saponified with 15 ml. of 0.5 N alcoholic potassium hydroxide according to the A. 0. A. C. Official Method.” The reaction mixture was then treated with 1 ml. of 6 N sulfuric acid and evaporated. After removal of the alcohol, the residue was boiled briefly with 20 ml. of 0.2 N sulfuric acid and the liberated fatty acids were extracted with 40 ml. of ether. The extract was washed with three 10-ml. portions of water and evaporated to dryness; yield 1.020 g. The combined aqueous layer and the washings were neutralized and evaporated to a dry residue which (14) E. C. Koo and P. S. King, I n d . Research ( C h i n a ) ,5, 137 (1936); C. A , , 30, 7887 (1936). (15) E. hZ. James in “Cotton Seed,” A. E. Bailey, ed., Interscience Publishers, Inc., New York, N. Y . , 1948. p. 706; (16) A product of the Floridin Co., Warren, Pa. (17) Association of Official Agricultural Chemists, “Official Methods of Analysis,” 7th ed., Washington, D. C., 1950, p. 435.

[CONTRIBUTION FROM

THE

2499

was extracted with three 20-ml. portions of absolute alcohol. Removal of the alcohol by evaporation under reduced pressure furnished crude glycerol; yield 123 mg. Conversion of Cotton Seed Oil and Fatty Acids to Barium Carbonate.-A sample of the oil (GU. 12 mg.) was weighed in a cup and oxidized with 0.5 g. of potassium iodate and 25 ml. of the Van Slyke-Folch reagent, by heating with a very small flame for 15 min. The resulting carbon dioxide was swept out with a current of nitrogen and was absorbed in a solution of half-saturated barium hydroxide containing 2 % barium chloride. The precipitate of barium carbonate was filtered on a fritted glass crucible and thoroughly washed with water, while avoiding any unnecessary expo5ure to the atmosphere. It was then dried and weighed. The barium carbonate obtained in this manner from 12.5 mg. of the CI4labeled cotton seed oil and 13.7 mg. of the fatty acids amounted to 157.2 and 175 mg., respectively. Under the employed conditions the amount of blank was negligible and a sample of commercial cotton seed oil containing 79.2% carbon provided a 98.3% yield of barium carbonate. Periodate Oxidation of Glycerol.-The above isolated glycerol (123 mg.) was dissolved in 5 ml. of 0.5 df sodium phosphate buffer a t pH 5.8 and oxidized with a solution of 0.65 g. of sodium metaperiodate in 10 ml. of water. The formaldehyde and formic acid produced were recovered as barium carbonate (349.6 and 174.6 mg., respectively) according to the method described by Eisenberg.lz Counting Methods.-The samples of barium carbonate were counted a t infinite thickness, using a preflush flow counter1* connected to a scaler,lg and compared with a standard sample.lO The samples were counted long enough to reduce the random counting error to 3 ~ 2 % . (18) Radiation Counter Laboratories, Inc., Skokie, Ill. (10) Nuclear Chicago, Chicago 10, Ill. ( 2 0 ) Obtained from Tracerlab. Inc., Boston 10, Mass.

COLUMBUS 10, OHIO

NORTHERN UTILIZATION RESEARCH BRANCH’]

Crystalline Methyl a-Isomaltoside and its Homologs Obtained by Synthetic Action of Dextransucrase BY R. W.

JONES,

ALLENEJEANES, C. S. STRINGER AND H. 51. TSUCHIYA RECEIVED OCTOBER 7, 1955

A new series of reference compounds, methyl a-isomaltoside (methyl 6-O-cr-~-glucopyranos~-l-a-~-glucopyrano~ide) and its homologs, have been synthesized by t h e action of t h e enzyme NRRL B-512F2 dextransucrase. The synthesis mixture was fractionated by chromatography on a carbon-Celite column. The products (D.P. 2 through 5) were isolated as pure crystalline compounds. They were identified and characterized by periodate oxidation, alkoxy1 analyses, Rfvalues, melting points, optical rotations and X-ray analyses.

The action of the enzyme, dextransucrase from the organism Leuconostoc mesenteroides NRRL B512F,2on sucrose has been used previously to obtain the high polym er dextran3r4and the disaccharides, (1) One of the Branches of the Agricultural Research Service, U.S. Department of Agriculture. Article not copyrighted. (2) Leuconostoc mesenteroides N R R L B-512F is a substrain of N R R L 8-512. The origin of the strain B-512 and the production of dextran from it were reported by Allene Jeanes, C. A. Wilham and J. C. hliers, J . Biol. Chem.. 176, 603 (1948). I n 1950, the B-512F substrain supplanted B-512 for all work at the Northern Regional Research Laboratory. Since that time. the dextran from this substrain has been designated inexactly as B-512 in numerous publications. However, in this article and hereafter, the correct designation of the substrain will be used. The dextrans from B-512 and B-512F appear to he identical. (3) H. M. Tsuchiya, H. J . Koepsell, J. Corman, G. Bryant, M. 0. Bogard, V. H. Feger and R. W. Jackson, J . Bact., 64, 521 (1952). (4) H. M . Tsuchiya, N. N. Hellman, H. J. Koepsell, J . Corman. C. S. Stringer, S. P. Rogovin. R.1. 0. Bogard, G. Bryant, V. H. Feger. C. A . Hoffman, F. R. Senti and R. W. Jackson, THISJ O U R N A L , 7 7 , 2412 (1955).

leucrose6 and isomaltulose.6 In these syntheses, the glycosyl acceptors are believed to have been sucrose or D-glucose and D-fructose, respectively. A number of other carbohydrates may serve as glycosyl acceptors,4,7 as has been demonstrated by means of paper chromatography. When maltose was used as alternate acceptor, the trisaccharide panose was prepared.8 This paper reports the use of methyl a-D-glUC0side as alternate acceptor in the synthesis of an homologous series of sugar derivatives. Transfer of the (5) F. H. Stodola. H. J. Koepsell and E. S. Sharpe, ibid., 74, 3203 (1952). (6) F. H. Stodola, E . S. Sharpe and H . J. Koepsell, Abstract P a p e r s , A m . Chem. Soc., 126, 5 D (1954). (7) H. J. Koepsell, H. M. Tsuchiya, N. N. Hellman. A. Kazenko, C. A. Hoffman, E. S. Sharpe and R . W. Jackson, J. B i d C‘heiir., 200, 793 (1053). ( 8 ) Mary Killey, R. J. Dimler and J . E. Cluskey, T m s J O U R N A L , 7 1 , 3315 (1955),

2500

R.IT.JONES, A. JEANES, C. S. STRINGER AND H. M. TSUCHIYA

Vol. 78

glucosyl moiety from sucrose, the donor substrate, to carbon six of methyl a-D-glucoside, the acceptor substrate, would yield methyl a-isomaltoside (methyl 6-0-a-D-glucopyranosyl a-D-glucopyranoside). The methyl isomaltoside thus formed, as well as products of subsequent enzymic action, may in turn act as acceptors with deposition of a glucosyl unit on their terminal carbon. In this manner, an homologous series of a- 1,B'-linked glucosides is obtained. Unless controlled, the enzymic action would result in the formation of polymeric material. In this study, the conditions were established to give optimum yields of low molecular weight products. The glycosides, having degrees of polymerization (D.P.) 2-5, were fractionated on carbonCeliteQand cellulose columns, crystallized and charac terized. Heretofore when an homologous series has been produced by enzyme action and isolated, the products have been sugars rather than sugar derivatives and have resulted from depolymeric rather than synthetic enzyme action.'O The enzymic synthesis reported here has the advantageous feature of yielding only one anomeric form, the a-configuration. Recently, Scott and Sentill have degraded B-512F dextran by methanolysis and separated the mixed a-P-anomeric methyl glycosides of D.P. 1-11 on a cellulose column. However, the products D.P. 2 through 6 were not crystalline. Methyl pisomaltoside has been obtained as an amorphous solid by Mrolfrom, et a1.I2

mole of sucrose and 4 X 104 dextransucrase units. T h e preparative mixture had a pH of 5.2 and was 0.01 M wit! respect t o acetate ion. T h e synthesis was conducted a t 4 for 23 h r . a t which time t h e reaction was complete as determined by reducing value measurement.3 T h e mixture was steamed at 100' for 15 minutes t o inactivate the enzyme. Composition of Reaction Mixture.-In order t o remove terial, the preparative enzymic mixture was (v./v,) methanol and centrifuged. The inn (about 2y0 of the total carbohydrate) contained nothing that moved on a paper chromatogram when 1-batanol-methyl Cellosolve-water (2: 1: 1) was employed a? developing solvent. Using the same solvent and ammoniacal 1056 silver nitrate as a spraying reagent, the methanolsoluble fraction TWS analyzed by quantitative chromatography. Fractionation of Reaction Mixture.-Fractionation via$ carried out 011 a column 8.0 X 33 cm. containing 330 g. of Norit S C and 2-IU g. of Celite 501. T h e carbon-Celite mixture was pretreated with 59; ethanol, packed in the column and conditioned with 5';c ethanol in the maimer previously described.1° A portion of t h e reaction mixture containing 33 g. of glucose equivalents, a i measured by quantitative arithroiie analysis,13in 150 m l . of 5'; ethanol, w a s added t o the column drippiiig under gravit).. Tlie rate of flow was gradually increased t o 800-900 ml./hr. by applying vacuum. The fractionation was follorwd b y semie color te;t.14 The ethanol coiicenwhen it apgeared t h a t tlie insin porit 1i~idbceii eluted. T h e percentagcs ,i7, , 8 , I O atid were then iiicreasecl b y .i..i' I.A s will be discussed in ;i later sec ion, in additio~it o the glycoside >cries a second grou!) of Iiomologs conitituting the isomalto?e seriei of reduciiig sug:irs wac: pre-eiit in small quantity in t h e reaction rnixturc. D-Fructoie, methyl a-D-glucode, tlie \light trace of sucrose displaced in the order methyl a-i~ornalto4tlc ', 6.methyl wi>oinctltoy llf%,methyl a-isoExperimental iitaose by 13..5(,% :tiit1 both methyl a-i,omaltopeii~Loside xntl Enzymic Production of the Methyl a-Isomaltoside Series. -The dextransucrase preparation employed was derived by 14-15?,; et1i:inol. Tlie fraction mere tii;carded beca from Leuconostoc mesenteroides hTRRL B-512F and was t h a t t!ie ~ ~reducin t; used t o study t h e factors influencing molecular weight dis- c , > n ~ ~ ~ oofi l ~ tribution of enzymatically synthesized dextran.4 The fractio!i> (U .P 2 - 4 ) containing the lca>t amounts of impuritie; w e r e rechromatograplied on carbo:i-Celite. T h e inimethyl a-D-glucoside was a commercial preparation with s purified on :I cel1ulo.e melting point of 161-163". T h e sucrose was table grade p u r e mcthyl a-i,omaltopentnc ,ide columii u-ing l-but'tnol--mctli~-l Ilomlve-water ( 2 : 1 : 1) cane sugar. I veri t . Test enzymic syntheses were conducted at 4' oii 5-ml. eniineralization and Crystallization of Glycosides.mixtures containing various amounts of methyl a-D-glUCoside and sucrose. T h e low reaction temperature was em- 'The effluent from the carhoii-Celite column contained ployed because i t favors t h e synthesis of low molecuhr soluble 4lic:ttc. Thi; was removed by repeatedly d i s d v i n g weight reaction product^.^ One ml. of prechilled dextran- the glyco5idc i n water, filtering and taking t o drynesc. sucrase preparation was added t o 4.0 ml. of prechilled, Other iiioiganic matter w t s removed from the D.P. 2 fractio,i b y Dowex 50 and Duolite -4-4 ion exchange resins, and buffered solutions of methyl a-D-glucoside and sucrose. T h e enzymic activity in the reaction mixtures w a i 30 dex- from the other glycosides by cry~tallizatioii of the carbotransucrase units3 per ml. T h e concentrations of the ac- 11ydr:tte. 'The glyco,ides were nllized by dis-olving the atnorceptor substrate in t h e test mixtures were 0.1, 0.2, 0.4 : L I I ~ in hot 17adding hot ctlianol or metlianol 1.0 M and the level of sucrose was 0.3 M . The s y i i t h e i e ~ p!ious pc~~vder .P. 2 glycozitle did not crystalwere allowed t o proceed for 23 hr., a t which time, as s h o w 1 atid cooling slon.ly. T lize from rnet1i:inol; it ii very soluble i i i cold ab>olutemcthb y past observations, all of t h e sucrose was utilized. A n : i n o l . Iiiitid cr-stallizntioii of I1.P. 4 :ind 5 \vas induced equal volume of methanol was then added t o each reaction cooliiig the aqueous alcoholic solution in Ilry Ice until a mixture t o inactivate t h e enzyme and t o precipitate any p-ecipitate formed and \mrining t o dis-olve tlie .olid. This high molecular weight dextran formed. T h e amounts of c>.cle \v:i> rer)e:ited iiritil acIequ:,te See( glycosides in t h e supernatant liquors mere determined by Identification of the Isomaltose Seri quantitative paper chromatography,'a using 1-hut:t:iol i.oiii.tlto,e serie- of D . P . = 77 !\-:I- f o methyl Cellosolve-water (2: 1: 1) as solvent. i t l e o f D . P . = i i - 1 from tlie ErLictioiiationOII On t h e basis of these analytical d a t a , t h e condition \w. lite coluiri!~.. Iwiiiciltotriow a n d i m m l t o t e t r a selected under which optimum results were obtained. a n d purified by Lirge-wale paper chromatogrnphy. one liter of a preparative enzymic reaction mixture w a ; bet I.i,maltohexaoie w:ti purified 011 i t cclluloie column. I t up. It contained 0.4 mole of methyl a-n-glucoside, 0.:3 w i i i iiut pructic;il t o 1,urify t h e i~om:tltr,i)entaose,bcc:iuse (9) The mention of firm names or trade products does not imply it IV:I, prewnt iri very irn:tll :tmi,ullt with a gre:tt quantity that they are endorsed or recommended by the U. S. Department uf of I1.P. 4 glyco?idc. LTtitlcr thc colldition~used, t h e first Agriculture over other firms or similar products n o t mentioned. meinbcr oi t h e iericy, i.onialti,ic, \\oulcl lie c:spected t o have (10) Allene Jeanes, C. A. Wilham, R. W.Jones, H. 51. Tsuchij-a and I,ceir eluted from tlic c:trboii along wit11 leucrose. A paper C. E. Rist, THISJ O U R X A L . 75, 5911 (1953). ci:roiii~ttc,gr:itnof t h e frxction affortletl cvitlerice of two com(11) T. A . Scott, Jr.. and 1'. K. Senti, hid 77, 38113 ( 1 9 i i ) . Ilo,i,Fiit., h i i t tlic 111iii~ir coii~titucrit\v:i\ i i t i t itleiitificd. I lie\c \lll,.t:uicc-~ \\-iiic!i trioveil O I I i);il)rra t t h e .;:imp r:itc ( 1 2 ) A I . 1.. TVIfrom,I.. \l' (.scorgcs iinrl I . 1. lliller, i R i C i , 71, I?,-, ( l ! i l ! l ) , I , ln,'lnl)cr., of t l i i , i\oiii:ilto,c i c . i . i r 5 , \\c.rc 1iIovcil t i l I)c ~ r c r >

(1:i)

c

I< ,I 1)1111lc.r,\\ C 5rliLccicr,i' I I 1 I tI p., I

[alZ5D

( c 2, water)a

O C .

1

2 3 3 4 4

EtOH 15.9 15.4 ... ... EtOH 11.7 11.3 MeOH ... ... EtOH 8.90 8.71 MeOH 3 EtOH 4.55 4.54 5 MeOH Calculated on dry weicht basis. Solvent: rotation.

1.07

o.oc

1.08 1.0' 1.04 0.25

110-1 11 232 140-146 162-167 192-193 161-164

+177.4 182.3 185.4 186.7 188.6 188.7

.13

.07

1-butanol-methyl Cellosolve-water (2 : 1 : 1).

ducing sugars by use of the chromatographic spray reagent 2,3,5-triphenyltetrazolium chloride.15J6 Their identity was established further from the RRr values of their partial acid hydrolysis products. Hydrolysis mas carried out on a steam-bath for 3 hours in 0.4 Nsulfuric acid. The acid was removed by neutralizing with saturated barium hydroxide and centrifuging. A paper chromatogram showed t h a t each hydrolyzate contained D-glucose and the expected members of the isomaltose series. I n addition, the isomaltotriose hydrolyzate contained a trace of reducing sugar apparently having a linkage other than cu-1,6'. When neutralized hydrolyzates were chromatographed and sprayed with ammoniacal silver nitrate, an unidentified yellow spot of X r near t h a t of methyl a-D-glucoside appeared. On drying and heating, it became purple. Analytical Determinations.-D .P. 2 samples were prepared for analysis by heating in Z'UGUO (Abderhalden) a t 79' and the other samples by heating at 137". No loss of weight occurred. T h e crystals were not hygroscopic under normal atmospheric conditions. They did prove t o be hygroscopic in 65% relative humidity a t 22'. Methoxyl content was determined by the Viebock and Schwappach method.'' These analyses showed the crystals t o be alcoholates. The alcohol was so tightly bound t h a t dissolving in water and taking to dryness in the Abderhalden at 79" did not eliminate all of it. Dry weights for all subsequent analyses were wlculated assuming exactly 0, 0.25 or 1 mole of alcohol per mole of glycoside (see Table I ) . Periodate oxidation analyses were carried out using a slight modification of the procedure of Rankin and Jeanes.I8 Fifty-milligram samples were weighed on a microbalance, and for the final aliquot 0.01 N iodine was used instead of the recommended 0.1 N iodine. Reaction appeared to be complete at 430 hr. Paper chromatographic techniques used have been described elsewhere.13,19

Results The amount of each component in the test and preparative reaction mixtures, determined by paper chromatography, is shown in Table 11. Physical constants, melting points, optical rotation and Rf values of crystalline products are listed in Table I. The purity and identity of the crystalline products were established by paper chromatography, periodate oxidation (Table 111) and alkoxyl analysis (Table I). Paper chromatography showed the D.P. 2, 3 and 5 glycosides to be pure. The D.P. 4 component contained a trace of isomaltopentaose. The Rf values, using 1-butanol-methyl Cellosolve-water ( 2 : 1: 1) as a developing solvent, are given in Table I. (15) J. W. White, Jr., Arch. Biochem. aizd B i o p k y s . , 39, 238 (1932). (16) W. E . Trevelyan, D. P. Proctor and J. S. Harrison, N o t w e . 166, 444 (1950). (17) E. P. Clark, J . Assoc. O f i r . Ap?,. C'iirinisfs, 1 5 , 136 (19x2). ( 1 8 ) J . C. Rankin a n d Allene Jeanes, Tms J O U R N A I . , ~ ~4435 , (1934). ( 1 9 ) Allme Jranrq, C . S. Wisr iind R . 1. T)imler, .tiiiii ( - h ~ i i i, 23, 41.5 (l