The Synthesis and the Enzymolysis of Methyl Glycosides of Malto-oligosaccharides' BY JOHN FI. PAZUK, JEAN M. MARSHAND TADAHIKO AN DO^ RECEIVED OCTOBER 16, 1958 The methyl a- and B-D-glycosides of nialto-oligosaccharitlcs were synthcsizctl cnzyniatically from cyclohexaainylosc and appropriate cosuhstrates. nietltyl a - D glucoside or methyl 8-wglucoside, by use of mocerans amylase. The glycosides of maltose, maltotriosc and maltotetraose have been isolated in pure form by chroinatographic proccdurcs. Methyl 8-Dmaltosidc. methyl a-D-nialtotrioside and methyl a-D-maltotetraoside have been obtained in crystalline form. Yeast maltase and fungal ainyloglucosidase hydrolyzed the methyl maltosides to n-glucose and methyl D-glucoside. Crystalline salivnry amylase hydrolyzed all of the a - D ( 1-r 4)-glucosidic bonds in the trioside and tetraoside but at varying rates and was without action on the bonds of the methyl maltosides.
Prior to the recent report of Peat, Whelan and J o r i e ~ ,the ~ synthesis of methyl a-D-maltoside and the methyl glycosides of higher niolecular weight malto-oligosaccharides had not been achieved. Methyl heptaacetyl a-D-maltoside had been prepared earlier by Inouye, et u L . , ~ by isomerization of the corresponding @-compoundwith the aid of antimony pentachloride catalyst. Dcacetylatiori of the a-compound should yield the inethyl a-D-maltoside. Although this procedure could be iised for the synthesis of the methyl glycoside of maltose, it is not readily applicable to the synthesis of glycosides of higher niolecular weight malto-oligosaccharides. Peat and associates3 prepared the methyl a-Dglycosides of maltose, maltotriose and maltotetraose in amorphous but chromatographically pure form by an enzymatic procedure. Methyl a-D-glucoside and amylopectin incubated with Denzymes were converted into a series of nonreducing methyl malto-oligosaccharides and into a series of reducing malto-oligosaccharides. The separation of the complex mixture of conipounds though eventually achieved was nevertheless quite tedious. In our laboratory the enzymatic synthcsis of the methyl glycoside series of maltooligosaccharides has been effected by iise of the coupling and redistribution reactions of mzccrans amylase. This method possesses certain advantages as pointed out below and has resulted in the preparation of some of the glycosides in crystalline form. Information on the course of hydrolysis of the methyl glycosides by yeast maltase, fungal amyloglucosidase and salivary amylase has bcen obtained and is also presented. Earlier studiesa have shown that rnaccrans amylase effects a transfer of the glucosyl units of cyclohexaamylose to methyl a-D-ghcoside to form niethyl a-D-iIialtohel)taoside which, in turn, via redistribution reactioiis,' is coiiverted into a hoinologous series of methyl glycosides. Sincc in the ( 1 ) Publislied with t h e approval of the Director as P a p a No. 010 Journal Series, Nebraska Agricultural Experiment Staticni. Slipported in part by a grant from t h e National Science 1'uiiuilat:mi. Presented before t h e Division of Biological Chemistry at the l33rrl National Meetiny of t h e American Chemical Society, San Francisco, Calif., April 13, 1858. ( 2 ) Fulbriyht exchange scholar sponsored. in part. by Takamine Laboratory, Division or Miles Laboratories, Inc., Clifton, N. J. (3) S. Peat. W. J . Whelan ;ind G. Jones, J . Chefn. SOL.,2100 (1957). (4) Y. Inouye. K. Oncdero. I. Karasawa. and Y. Nishisawa. J . A g r . Chent. soc. ( J a p a n ) , 26, G 3 1 (1032). (5) S. Peat, W. J. Wbelari and W .R , Ilce,. J. < ' / I P U I . Suc.. .I4 (13SG). ( t i ) 11 l:rcnLli, h l . I..Leviiic, I C . Nnrbcrx, 1'. Nordin, J . 11. l'.uilr a n d C;. h l . Wilil. 'L'iiis JO(IIINAI.. 76, 2;iX7 (1!131). (7) E. h'urbcrg aud L). l ~ i ~ i i ct lb~r J, , 72, 1 3 2 ( l ~ ~ i l ) .
m c c r a n s digest no reducing co-substrate is prcscnt, the reducing series of malto-oligosaccharides is not produced and the difficulty of separating the reducing and non-reducing compounds as experienced by Peat, et a1.,3is not encountered. The individual methyl glycosides are resolved easily by chroniatography on paper or on cellulose powder columns with a n-butyl alcohol-ethyl alcohol-water s y s t c ~ r ~ ( 4 : l : l by volume).8 The methyl 8-D-glycosides of the rnalto-oligosaccharides also have been prepared by use of the mcerans enzyme. I n this case the cosubstrate employed was methyl P-D-glucoside. From a digest of methyl P-D-ghcoside and cyclohexaamylose with m c e r a n s amylase, methyl P-Dglucoside and methyl @-D-mdtosidc were isolated in crystalline forni and methyl @-D-rnaltotrioside and methyl P-D-maltotetraoside in amorphous form. The apparent Rf values for the P-series of compounds, for the a-series and for the unsubstituted nialto-oligosaccharides are recorded in Table I. 1'Alll,E
AITARIJN L' /\'f
vAl.UliS"
TIIEIR
1
(?lc ~ ~ A l . ~ L ' O - O ; l c u f cry>talliue inetliyl I? JJ i u . d L ! ~ d c , : ~ i i dS l n t e
May 5, 19.59
h l E T l l Y L GLUCOSIDES OF h ~ A L T ~ - ~ l ~ I G O S A C C l l A R l D ~ ~
2171
of the enzyme containing 28 SKB unitsu per ml. glucoside but did not hydrolyze the methyl gluco- solution was employed for hydrolyzing the triosides and a solution sidic bond of the compound, containing 9.0 SKB units per ml. for hydrolyzing the tetra@ The specific rotation and the melting point of side. Yeast maltase of Saccharomyces ccrcviskc was prethe crystalline methyl a-D-maltotrioside were pared by a published procedure1@with slight modscations. +202O and 145-147'. Crystalline salivary amyl- Amyloglucosidase of Aspergillus triger was obtained in a highly purified form by a chromatographic procedure emase hydrolyzed the maltotrioside in a pattern very ploying cellulose ion exchange material.17 The details of similar to that for the hydrolysis of maltotriose,@ the preparation of this enzyme will be published elsewhere. Preparation of the Methyl Glycosides.-Four grams of ;.e., bond 110 of the trioside and triose were hydrolyzed a t a faster rate than bond 2. Use has methyl a-mglucoside and 4 g. of cyclohexaamylosel* were dissolved in 80 ml. of water and mixed with 20 ml. of been made of the difference in rate of hydrolysis mceruns amylase solution (1.4 units per tnl.), Aliquots of the two bonds in the trioside to obtain two of the digest were examined for reaction products by paper hydrolytic fragments which aided in the character- chromatography (solvent system, n-butyl alcohol-ethyl 4: 1: 1 by volume). The non-reducmg ization of the non-reducing compound as methyl alcohol-water, compounds were detected with permanganate-periodate a-D-maltotrioside. These were methyl a-D-glUC0- rmgentm and reducing compounds with copper sulfate and side and maltose, the former being identified by its phosphomolybdic acid reagents." In the macerans digest melting point and X-ray diffraction pattern and the there was produced a series of non-reducing compounds latter by its specific rotation and X-ray diffraction (Rfvalues shown in Table I) but no reducing compounds. A plot of the Rfvalues in Table I wrsus degree of p o l y m d pattern. zation as outlined by French and Wild" resulted in a straight The methyl a-D-maltotetraoside crystallized line relationship. As pointed out by French and Wild such as a dihydrate with a specific rotation of +201° a linear relationship indicates strongly that structural irand a melting point of 170-172'. Paper chroma- regularities do not exist in the series of compounds. digest of non-reducing compounds was concentrated tographic evidence indicates that the methyl a - ~ -t o The a volume of approximately IO ml., mixed with 5 g. of maltotetraoside was hydrolyzed by salivary amylase washed cellulose powder,%and taken to dryness in a Vacuum in a manner quite similar to maltotetraose." desiccator. The material was stirred with a Small volume The major products of hydrolysis of the tetraoside of a solution of n-butyl alcohol-ethyl alcohol-water (4: 1:1 by volume) and introduced on a cellulose column (70 g.; 300 were maltose and methyl a-D-maltoside, tentatively mm. )( 40 mm.) which had been thoroughly Washed with identified by paper chromatography. Small the above solvent mixture. The development of the amounts of reducing compounds which appear to column was continued with the n-butyl alcohol-ethyl a h be glucose and maltotriose also were detected hol-water solvent until 250 fractions (10 ml.) were collect+. on the chromatograms. The concentration of Aliquots of these fractions were heated with diphaylamme reagent11 t o locate the tubes containing carbohydrates and maltotriose in the hydrolysate was greater than the identities of the compounds in the various fractions Were that of glucose indicating that the terminal bond then determined by qualitative paper chromatography. 1'O was hydrolyzed a t a somewhat faster rate than Pure methyl a-D-glucoside was present in fractions 21 t o 38, the terminal bond 3. On the basis of the chroma- pure methyl a-mmaltosidc in fractions 42 t o 78, pure methyl tographic evidence the rate constants for the hy- u-smaltotrioside in fractions 90-135 and pure methyl a-Dmaltotetraoside in fractions 182-195. The fractions condrolysis of the terminal bonds of the tetraoside taining the individual pure components were combined, are much smaller than the constant for the hy- concentrated under vacuum t o a small volume and allowed drolysis of the central bond. These patterns of t o stand a t room temperature. Methyl a-wglucoside, hydrolysis of the maltotrioside and maltotetraoside methyl a-smaltotrioside, methyl a-D-maltotetraoside were in crystalline form from the solutions. Methyl by salivary amylase differ from those reported by obtained a-smaltoside came out of solution as a sirup which after Peat, et Apparently, in the experiments of the evaporation of the solvent was dried in a vacuum desiccator investigators mentioned above, the slow rate of to constant weight. The yields of compounds were as hydrolysis of the terminal bonds of the tetraoside follows: methyl a-D-ghcoside, 0.6 g.; methyl a-D-makoside, 0.5 g.; methyl a-wmaltotrioside, 0.4 g.; methyl a - ~ and bond 1of the trioside escaped detection. maltotetraoside 0.3 g. Enzymatic hydrolysis of methyl 8-D-maltoside A similar procedure was used for preparing the methyl by yeast maltase and fungal amyloglucosidase oc- 8-mglycoside series of maltosligosaccharide. I n this curred by routes similar to those for the methyl experiment 4 g. of methyl 8-mglucoside was used as the cosubstrate. The chromatographic behavior of the methyl a-D-makoside. Likewise, salivary amylase hy- 8-series of compounds was quite similar to that of the adrolyzed methyl 8-D-maltotrioside and methyl series (see Table I). The methyl @-D-glycosidesof mglucose 8-D-maltotetraoside by patterns very similar to and maltose were obtained in crystalline form while those of maltotriose and maltotetraose were obtained in amorphous those for the a-series of compounds. form. Methyl a-D-Glucoside.-The compound with Rf value Experimental
0.86 isolated from the enzymatic digest of methyl a-D-glucoEnqmes.--Muceruns amylase was prepared from BaciUus side and cyclohexaamylose prcfved t o be methyl a-D-glucomceruns by methods previously described.l* The enzyme side. T h e melting point of the compound was 167' and sample used in this study posessed 1.4 unitsla of activity the mixed melting point with an authentic sample of methyl per ml. It was free of hydrolytic activity as indicated by a-wglucoside was 166". The specific rotation of +157" the absence of reducing products in digests of methyl a - ~ - agrees with the literature value for methyl-a-D-glucoside. glucoside and cyclohexaamylose. Crystalline salivary amyl- An X-ray diffraction pattern" of the material isolated from =el4 was the gift of Dr. Jytte Muus. I n our experiments, a (9) J. H. Pazur and T. Budovich, Science, 141, 702 (1955). (10) The glucosidic bonds in the oligosaccharides are numbered consecutively beginning with the a-o-(l-4)-bond joining the reducing glucose unit (or methyl glucoside residue) to the remainder of the molecule. (11) J. H. Pazur, J . B i d . Chcm., 408, 75 (1963). (12) D. French, M. L. Levine, J. H. Pazur and E. Norberg, Tar9 JOURNAL, 71, 353 (1949). (13) E. B. Tilden and C. S. Hudson, J . Bad.. 43, 527 (1942). (14) J. Muus. Compt. rend. trap. Lab. Carlsbrrp, 98, 317 (1953).
(15) R. M. Sandstedt, E. Kneen and hl. J. Blish, Cereal Chem., 16, 712 (1939). (16) R. WillstHtter and E. Bamann, 2. phyriol. Chem., 161, 242 (1926). (17) M. B. Rhodes, P. R. Azari and R. E. Feeney, 1. B i d . Chem., 430, 399 (1958). (18) R. U. Lemieux and H. F. Bauer, Anal. Chcm.. 46, 920 (1954). (19) D. French and G. M. Wild, THISJOURNAL, 76, 2612 (1953). (20) Solka-Floc BW-200, Brown Co., Berlin, N. H. (21) All X-ray patterns were obtained by Professor W.C. Robisoo, UniveSsity of Nebraska Instrumentation Laboratory, Lincoln, Neb.
2172
JOHN
H. PAZUR, JEAN M. MARSHAND TADAHIKO ANDO
the enzymatic digest was identical to t h e pattern of methyl a-D-glucoside. Methyl a+Maltoside.-The compound with Rt value 0.54 was non-reducing and possessed a specific rotation of +184". In a n acid hydrolysate of t h e compound (0.05 N HC1 at 100' for 1 hour) two reducing fragments, one with Rf value identical t o t h a t of D-glucose and t h e other with Rr value identical t o t h a t of maltose were detected b y paper chromatography. The concentration of maltose in the hydrolysate was considerably lower t h a n that of D-glucose even m the early stages of hydrolysis, indicating t h a t t h e a+-( 1+4)-glucosidic bond in the compound is hydrolyzed more rapidly t h a n the methyl glucosidic bond. T h e compound (0.05 9.) dissolved in 0.5 ml. of water was treated with 0.5 ml. of a solution of fungal amyloglucosidase. After a 6-hour incubation period the sample was heated at 100' for 5 minutes t o inactivate the enzyme and the compounds in t h e hydrolysate were separated on cellulose powder column (7 g.; 230 mm. X 12 mm.) with the n-butyl alcohol-ethyl alcohol-water solvent system. From t h e combined eluates in tubes 23-32, there was obtained a crystalline compound with m.p. 166' and with X-ray diffraction pattern identical t o t h a t of methyl a-D-glucoside. From the eluates in tubes 38-50 there was obtained a crystalline reducing compound with specific rotation of +53". T h e compound yielded an X-ray diffraction pattern identical to t h a t of D-glucose. Methyl a-D-Maltotnoside.-The crystalline non-reducing compound with Rfvalue 0.22 and specific rotation of $202' is evidently methyl a-D-maltotrioside. T h e melting point of the compound was 145-147'. Anal. Calcd. for C19H34Ol6: C, 44.05; H, 6.56. Found: C , 43.77; H , 6.90. The X-ray diffraction data on the compound were: 12.01=70,'3 9.64-60, 7.93-50, 7.12-80, 6.47-80, 5.45-50, 5.19-60,4230-50,4.56-40,4.3740,4.08-100,3.77 (double)80, 3.58-70,3.42-50, 3.19-30, 3.08-30,2.9&30, 2.81-20. In preliminary experiments, i t was found t h a t the methyl a-D-maltotrioside was hydrolyzed by crystalline salivary amylase t o D-glucose, maltose, methyl a-D-glucoside and triethyl a-D-nmltoside. This action pattern on methyl inaltotrioside is very similar t o t h a t observed earlier for the action of salivary amylase on ma1totriose.Q A sample of 0.12 g. of the trioside in 2.0 ml. of water was treated with 2.0 ml. of salivary amylase solution (28 SKB units per mt.). After a 12-hour incubation period a t room temperature, the products in the hydrolysate were separated on a cellulose powder column (20 g.; 220 mm. X 22 mm.) with n-butyl alcohol-ethyl alcohol-water (4: 1: 1 by volume) solvent. Paper chromatograms of the eluate showed t h a t tubes 25 t o 36 contained a compound with R, value of methyl a-D-glUC0side. The samples from these tubes were combined, concentrated to a small volume and allowed to stand at room temperature. On further evaporation of the solvent a crystalline conipound with m.p. 166" and with a n X-ray diffraction pattern identical t o that of methyl a-D-glucoside was obtained. The solution from tubes 91 to 135 contained ;i reducing compound with Rf value typical for maltose. There was obtained from the combined solution from these tubes a crystalline product with a specific rotation of +129" and with a n X-ray diffraction pattern identical to t h a t of maltose. -~
Vol. 81
Methyl a-D-Maltotetraoside .-As indicated in Table I, the Rf value of the fourth member of t h e series of compounds was 0.06. The specSc rotation of +201' for the crystalline product is lower than t h a t reported by Peat, et al.,a indicating t h a t t h e compound contains some solvent of crystallization. Quantitative values for t h e elemental composition of the compound agree with the values for a dihydrate of the compound. Anal. Calcd. for C25H1,02~.2H20: C, 41.92; H , 6.72. Found: C , 41.85; H, 6.68. T h e specific rotation calculated on the ailhydrous basis is +212' arid is in good agreement with the literattfre va1ue.l The X-ray diffraction data on the crystalline compound were: 8.54**-7OZ3,7.8660, 7.34-10, 6.47-20, 5.87 (double)80, 5.10-100, 4.87-80, 4.48-60, 4.05-50, 3.85-10, 3.63-80, 3.3540, 3.14-50, 3.02-70, 2.85-60, 2.76-20, 2.68-30, 2 . 6 2 4 0 , 2.56- 10, 2.49-20. A sample of 0.009 g. of the compound tii.isolved i n 0.1 ml. of water was treated with 0.1 ml. of s:ilivary amylase (9.0 S K B units per nil.) for 3 hours. .41ialysis of the 0- and 3-hour aliquots b y paper chromatography showed the presence of three reducing cornpounds in the 3-Iir. digest. The one present in highest concentration had an Rf value of 0.42 ( K I value of maltose under the same conditions 0.42) and the other two present only in trace amounts with Rr values of 0.68 and 0.35 ( R fvalues of D-glucose and makotriose, 0.67 and 0.35, respectively). Methyl fl-D-Glucoside.-The fsstrst rnoving compound in the enzymatic digest of methyl fl-n-glucoside and cyclohexaamylose proved to be methyl $D-glucuside. The m.p. of this compound was 105'; mixed m.p. with authentic methyl 8-D-glucoside was 105'. The specific rotation of -33' a n d the data for X-ray diffrwti.,n diagram of the compound agree with the values for methyl 3-n-glucoside. Methyl-2-D-Maltoside. -Tlie second nieniber of the 8series o f compounds also was obtained in crystalline form; ii1.p. 110' itnd specific rotation +84O. Thcse vslues agree with the litcrature values for methyl B-~-inaltoside.*~The X-ray diffraction pattern of this compaund was identical tu that obtained for pure mctliyl &n-rnaltositle.b The compound (0.01 9.1 dissolved in 0.1 nil. of iv:itcf was ixcubated with 0.1 ml. of yeast rn:ilt;tsc si)lution at 30" for 0, 1, 2, 4 and 6 hours. I n thc 1-hr. Iiydrolysate there was detectcd by paper chromatography a reducing compourld :irid a non-reducing compound which w e , in all likelihood. wglucose dnd methyl pD-glucoside. T h conccntratiori of thcse compounds increased progressivcly up t o the 6-hr. sample. Methyl 6-n-Maltotrioside and Methyl p-D-Maltotetraoside. --The third and fourth nicrnbers uf the B-series were a h ul)tairicd in chroin~tographicallyp u r e form; Kf valum for these compounds were (1.24 and 0.07, rcspectivcly. 0 1 1 the basis of chr,matogr;tphic behavior anti the rxperience in the prcparaiion and characterization of the a-series of coinpounds these conipounils are probal)l>. methyl 8-Dnulrutrioside and inethyl ~ - r ~ - i i i , t l t u t c t r a u s i ~ IRcduciiig ~. iii hydrulyfrngmcnts c!c.tcctrtd h y paper chrclri~;iti~grirpIiy s:ites of these coinpounds with sctli\':iry ani>'I.ts'et bc.c.ri ( ~ 1 ) raincd. LI :.L~\ll.x,s 1: I ) . 811
~
(22) Interplanar spacings, b., CuKa radiations. ( 2 3 ) Relative intensities on the basis of 100 for the strongest line.
? I < 'I 5 c l ~ , c h ,1: .I \Yilson, J r , and C. 5 . It:icI.uu, 'I HI:, 64, 2871 ,tY121.
J U C R ~ < I ,
