- Email: [email protected]
Acetolysis of Pichia pastoris IFO 0948 strain mannan containing α-1,2 and β-1,2 linkages using acetolysis medium of low sulfuric acid concentration
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 245, No. 2, March, pp. 494-503,1986
Acetolysis of Pichia pastoris IF0 0948 Strain Mannan Containing a-l,2 and ,f3-1,2 Linkages Using Acetolysis Medium of Low Sulfuric Acid Concentration HIDEMITSU second Department
KOBAYASHI, of Hygienic
NOBUYUKI SHIBATA,
AND
Ch.emist?y, Tohoku College of Pharmacy,
SHIGEO SUZUKI’ S&i,
Miyagi
983, Japan
Received June 21, 1985, and in revised form October 22,1985
To obtain manno-oligosaccharides containing @-1,2-linked nonreducing terminal groups from the mannan of Pichia pastoris IF0 0948 strain by acetolysis, an attempt was made to establish the reaction conditions under which cleavage of the a-1,6 linkage took place preferentially leaving manno-oligosaccharides composed largely of p-1,2 linkages. By the action of an ordinary acetolysis medium, a 10/10/l (v/v) mixture of acetic anhydride, acetic acid, and sulfuric acid at 40°C for 13 h or at 25°C for 120 h, the Oacetyl derivative of this mannan gave mannose, mannobiose, mannotriose, and mannopentaose. However, treatment of the same O-acetyl mannan with a 50/50/l (v/v) acetolysis medium at 40°C for 15 h gave a mannotetraose in addition to mannose, mannobiose, mannotriose, and mannopentaose. Use of a 100/100/l (v/v) acetolysis medium at 40°C for 36 h gave a more satisfactory result, a mixture of oligosaccharides, from mannose to mannopentaose, which contained more mannotetraose than mannopentaose. Because both mannotetraose and mannopentaose contained a-1,2 and p-1,2 linkages, it was concluded that an acetolysis medium containing a low concentration of sulfuric acid, up to 0.5% (v/v), facilitates the preferential cleavage of the a-1,6 linkage, leaving manno-oligosaccharides containing the B-1,2 linkage which was found to be labile to the action of the 10/10/l (v/v) acetolysis medium. o 19s Academic PW., IN.
It is well known that cell wall mannans comprise the principal antigens of the parent yeast cells (1, 2), and that controlled acetolysis provides a convenient method for the analysis of immunochemically specific sites corresponding to the branching moieties of the parent mannans (3,4) and for determining the chemical structure of the intact mannans (5). Gorin and Perlin (6) first applied acetolysis to the structural study of yeast mannans, and were able to isolate Manpal - 2Man.’ Thereafter, 1 To whom all correspondence should be addressed. ’ Abbreviations used: PMR, proton magnetic resonance; Manpal + 2Manppl 2Manpfil 2Manpal - 2Man, ~~-D-mSnnOpyranOSyl-(l 2)~~-D-mannOpyrSnOSyl-(l
nosyl-(1
-
-
2)-~~-D-mannOpyra-
2)-C&x-D-mannopyranosyl-(1
0003-9861/86 $3.00 Copyright All righta
Q 1986 by Academic Press, Inc. of reproduction in any form reserved.
-
2)-D-
many papers were published using acetolysis to study various yeast mannans (4,5, 7-10). Some yeast mannans, however, were shown to contain @linked mannose residues from the results of proton magnetic resonance (PMR) studies by Gorin and Spencer (11). Structural information obmannose; Man@1 - 2Manp@l2Manpal2Man, O-B-D-mannopyranosyl-(1 + 2)-@@-D-mannopyra- 2)-Dnosyl-(1 - 2)-&x-D-mannopyranosyl-(1 mannose; Man@1 - 2Manpal - BMan, o-b-D-mannopyranosyl-(1 - 2)-&x-D-mannopyranosyl-(l 2)D-mannose; Manpal - 2Manpal - BMan, &Y-Dmannopyranosyl-(1 - 2)-Oa-D-mannopyranosyl-(l + 2)-D-IIUmnOSe; Manpal - SMan, O-a-D-mannopyranosyl-(1 - 2)-D-mannose; Manp~l - BMan, o-@-Dmannopyranosyl-(1 - 2)-D-mannose; Manpal -+ GMan, ~cY-D-mSnnOpyrSnOSyl-(1-+ 6)-D-mannose. 494
ACETOLYSIS
OF b-1,2 LINKAGE-CONTAINING
MANNAN
495
tained by acetolysis on these mannans was was made by means of a Shimadzu Chromatopac-ElA limited in another report by Gorin and his microcomputer. Materials. The P. pastoris IF0 0948 strain was obco-workers (8). The results of our investitained from Institute for Fermentation, Osaka, Japan. gation on acetolysis conditions for the A column packing for gel-filtration chromatography, mannan from Pichia pastwis IF0 0948 Toyopearl HW-40 (superfine), fractionation range of strain, which show a PMR pattern closely 100 to 7000 Da, was purchased from Toyo Soda Manidentical to that of the P. pastoris PRL 63- ufacturing Company, Ltd., Tokyo, Japan. 208 strain mannan published by Gorin et Cultivation of P. pastoris IF0 0948 strain The P. al (8), showed discrepancies with those ob- pastoris IF0 0948 strain was grown in 500-ml flasks tained by these workers (8). For example, containing a liquid medium described by Ishitani et the chemical structure of this mannan as al (17) on a reciprocal shaker at 28°C. The composition proposed by these workers (8) contains only of the liquid medium was as follows: NH&l, 4.0 g; one kind of branching moiety correspond- KaHPO,, 2.0 g; KHaPO,, 2.0 g; MgSOd.?HaO, 0.5 g; corn steep liquor (Sigma Chemical Co., St. Louis, MO.), ing to a pentaose, Manpal - 2ManpPl 2.0 g; glucose, 20 g; in 1000 ml of water (pH 6.0). The 2Manp/31 - 2Manpal - BMan, although cells were harvested after 48 h, washed three times many yeast mannans were found to contain with saline by centrifugation, and dehydrated with a branching moieties of different length (5). large volume of acetone. In this paper, we describe a modified acPreparation of the mannan of P. pastor6 IF0 09.48 etolysis method with milder conditions for strain This was completed in accordance with the previous description for the isolation of a X cerevisiae the isolation of manno-oligosaccharides wild-type strain (18) by means of Fehling solution. composed predominantly of fi-1,2-linked Yield of the mannan was 12.5% on weight basis of the mannose residues including Manpal 2Manp@l - 2Manpcvl - 2Man from the acetone-dried whole cells. Acetolysis of mannan, Prior to acetolysis, the manmannan of P. pastoris IF0 0948 strain. nan was converted into its 0-acetyl derivative in acMATERIALS
AND
METHODS
General. Total carbohydrate and total phosphate were determined by the phenol-sulfuric acid method (12) and by the Ames-Dubin method (13), respectively. Total protein was determined by the Folin method of Lowry et al (14) using bovine serum albumin as the standard. Precoated silica gel plates (Kiesel Gel, without fluorescent indicator, 0.25-mm thickness, 20 X 20 cm, Merck, Darmstadt, West Germany) were used for thin-layer chromatography. The solvent used for thin-layer chromatography was a 5/3/2 (v/v) mixture of 1-butanol-ethanol-water. Sugars on the plates were detected by spraying with anisaldehyde reagent (15). Specific rotations were measured using an Applied Electric automatic polarimeter (c = 1.0, 1= 1.0, water). PMR spectra of the H-l region of mannan and mannooligosaccharides were recorded using a JEOL JNMFX 100 spectrometer in DsO solution at 70°C in accordance with the description by Gorin and Spencer (11). Sugar component analysis of the mannan of P. pastoris IF0 0948 strain was conducted in accordance with the description by Okubo et al (16) on that of the mannan of a Saccharomyces cerevi.siae wild type strain prepared by means of cetyltrimethylammonium bromide. Gas chromatography of Omethyl-0-acetyl mannitols was conducted on a glass column (5 mm X 150 cm) containing 3% OV-210 on Supelcoport (100 to 200 mesh) at 185°C using N2 as the carrier gas at a flow rate of 20 ml/min. Conversion of the peak dimensions into molar ratios of the sugar derivatives
cordance with the description by Okubo and Suzuki (19). Namely, mannan, 100 mg, and anhydrous formamide, 5 ml, contained in a 50-ml round-bottomed flask with a glass stopper were heated in a boiling water bath until the solid mass was dissolved, and the resultant solution was cooled to room temperature. To the solution was added a l/l (v/v) mixture of (CH&O)aO and anhydrous pyridine, 20 ml, at once, and the clear solution was kept at 40°C for 12 h. Then the reaction mixture was concentrated in nacho to a thick syrup, and the rest of the volatile materials was thoroughly removed by means of an oil diffusion pump to leave the 0-acetylated mannan as a hard glass in quantitative yield. Then the Oacetylated mannan was subjected to acetolysis under eight different conditions, designated Conditions Ia, Ib, IIa, IIb, IIc, IIIa, IIIb, and 111~ as follows: In Condition Ia, the 0-acetyl mannan was dissolved in a 10/10/l (v/v) mixture of (CH&O),O, CHaCOOH, and HaSO,, 10 ml, and the resultant solution was kept at 40°C for 13 h. This condition was the same as that described by Kocourek and Ballou (5). After termination of the reaction by the addition of pyridine, 2 ml, the mixture was concentrated in vucuo to a thick syrup, and dissolved in CHCls, 10 ml. Then the solution was shaken with water, each 15 ml, three times, and dried over anhydrous NaaSO,. After filtration, the solvent was evaporated in vocv.~ to dryness; then trace of CHCls was removed by codistillation with toluene, 5 ml. The residue was dissolved in anhydrous CHaOH, 2.5 ml, and 1 M methanol solution of CH80Na was added dropwise until the precipitation of de-O-acet-
496
KOBAYASHI,
SHIBATA,
ylation product started. After leaving for 30 min, the mixture was neutralized with 50% CH&OOH, and evaporated in vacua to dryness to yield a mixture of manno-oligosaccharides containing small amounts of CH&OONa. In Condition Ib, the composition of the acetolysis medium was the same as that of Condition Ia, while the reaction temperature and period were 25°C and 120 h, respectively. This condition was exactly identical to that for the structural study of the mannan of P. pastoris PRL 63-208 strain reported by Gorin et al. (8). Conditions IIa, -b, and -c involved treatment of the 0-acetyl mannan with a 50/50/l (v/v) mixture of (CH&O)zO, CI-I&OOH, and H&SO4 at 40°C for three different periods-3,6, and 15 h, respectively-in accordance with the description by Okubo and Suzuki (19) on a controlled acetolysis study of the mannan of a S. cerevisioe wild-type strain. Conditions IIIa, -b, and -c involved the use of an acetolysis medium of lower HzSOl concentration, a 109/100/l (v/v) mixture of (CI-I&O)zO, CH&OOH, and HzSO, at 40°C for three different reaction periods, 12, 36, and 120 h, which were designated Conditions IIIa, -b, and -c, respectively. The de-0-acetylated oligosaccharide mixture was dissolved in water, 2.0 ml, and was applied to a column of Toyopearl HW-40 (2.5 X 100 cm). Elution was made with water (0.25 ml/min), and aliquots of the eluates, each 15 ~1, were assayed for carbohydrate content by the phenol-sulfuric acid method (12). Eluates corresponding to each peak were combined, concentrated in w~1cuo to a small volume, and assayed for the homogeneity of the oligosaccharides contained in the solution by thin-layer chromatography. To remove small amounts of contaminated oligosaccharides of lower or higher molecular weight, the solution was rechromatographed on the same column of Toyopearl HW-40, and the eluates containing a homogeneous oligosaccharide were combined and lyophilized after concentration in vacua. Meth&&m an&s&. Methylation of manno-oligosaccharides, each 5 mg, obtained by gel filtration of the acetolysis products was conducted by the Hakomori method (20), and the resultant per-O-methylated manno-oligosaccharides were converted into a mixture of 0-methyl-0-acetyl mannitols according to the description by Lindberg (21).
AND
SUZUKI
resultant solution was kept at 40% At intervals of 1, 3, 6, 12, and 24 h, aliquots of solution, each 5 ml, were withdrawn, and were evaporated in vacua after addition of pyridine, 1.0 ml. The thick syrup was dissolved in CHCl,, 10 ml, and washed with water, 100 ml. The CHCh, layer was separated and dried over anhydrous Na#O,. After filtration, the solution was evaporated in vacua to dryness, and the residue was dissolved in anhydrous CHzOH, 3 ml. To the solution was added a few drops of 1 M methanolic CHsONa solution, and the mixture was left at room temperature until the precipitation of free sugars was accomplished. Then the mixture was neutralized with 50% CHaCOOH, and evaporated in v-0 to dryness. The residue was dissolved in water, 10 ml, and was deionized by the addition of a mixture of Amberlite IR-120 (H+) and Amberlite IR-410 (OH-) resins. The supernatant was collected by filtration, and the resin mixture was thoroughly washed with water. Then the combined filtrate and washing were concentrated in vaouo to dryness. After dissolving in a 68132 (v/v) mixture of CHsCN and water, 200 ~1, the solution was analyzed by high-performance liquid chromatography using an Erma ERC-8710 liquid chromatograph equipped with a column of ERC-NH-11’71,6 X 200 mm, and a differential refractometer. Elution was made with the same mixture of CHaCN and water at 100 kg/cm’, and the eluate wae monitored for the amounts of unchanged and degraded oligosaccharides. Under the chromatographic conditions, the relative retention times of mannose, mannobiose, mannotriose, mannotetraose, and mannopentaose were 1.00, 1.16, 1.48, 1.81, and 2.06, respectively. Degradation ratios of the starting mannotetraose or mannopentaose at five interval periods were computed by the following formula: Degradation ratio (a)= A/(A + B) X 100, where A represents sum of the peak dimensions of the degradation products including lower oligosaccharides than each corresponding parent oligosaccharide and mannose, and B indicates the peak dimension of unchanged parent oligosaccharide. It should be noted that a 100% degradation means the disappearance of each starting manno-oligosaccharide, tetraose, or pentaose in the reaction mixture, and not complete degradation to mannose. RESULTS
Determination of acetolytic rates of mmnotetraose and mannqpentaose in aatolysis medium of a lO/lO/ 1 (v/v) mixture of @H&O)&, CHJOOH, and HSO,.
Properties of the mannan of P. passtoris IF0 0948 strain. The results of chemical
Each oligosaccharide, 20 mg, was dissolved in 5 ml of anhydrous formamide in a 50-ml glass-stoppered, round-bottomed flask. To the solution was added a 11 1 (v/v) mixture of (CH&O)zO and anhydrous pyridine, 10 ml, on a boiling water bath until the solid mass dissolved. Then the clear solution was evaporated in vacua to dryness using an oil diffusion pump. The residue was dissolved in a mixture of (CH&O)zO, CH&OOH, and H&SO,, 10/10/l (v/v) ratio, and the
analyses of the mannan of P. pastoris IF0 0948 strain are shown in Table I. The sugar component analysis of the mannan revealed that mannose was the major sugar; the presence of a small amount of glucose was also evident. The mannan contained more than 90% total carbohydrate, 3.0% protein, and trace amounts of phosphate.
ACETOLYSIS
OF o-1,2 LINKAGE-CONTAINING TABLE
497
MANNAN
I
CHEMICAL COMPOSITIONOF THE MANNAN OBTAINED FROM P. pastoris
Total carbohydrate
Molar ratio of sugar eomponentb
(%I”
Mannose
Glucose
93.0
97.0
3.0
Total protein (%I”
Total phosphate (%ld
l@
Trace
f14.8
3.0
(“)”
a Determined by the phenol-sulfuric acid method (12). b Determined by gas chromatography by converting the component sugars of each acid-hydrolyzed into the corresponding aiditol acetates (16). ‘Determined bv the Folin method of Lowry et al. (14). d Determined by the Ames-Dubin method (13). BIn water, c, 1.0; 1, 1.0.
Therefore, this mannan was expected to give very small amounts of phosphatecontaining oligosaccharide(s) eluted in the void-volume region in the acetolysis fingerprint of its acetolysate obtained by gel chromatography as observed by a few workers on several yeast mannans (l&22, 23). This mannan also gave a PMR pattern closely identical to that of the P. pastoris PRL 63-208 strain mannan, demonstrating the existence of large amounts of (u-1,2and p-1,2 linkages corresponding to chemical shifts of 5.09 to 5.38 and 4.87 ppm in the PMR pattern of this mannan, respectively (Fig. 1). However, it should be noted that this PMR pattern is not consistent with the chemical structure of P. pastoris PRL 63208 strain proposed by Gorin et al. (8), because it was shown to contain a single kind of branching moiety corresponding to a mannopentaose residue, Manpcvl 2ManpPl - 2ManpBl - 2Manpal - 2Man. The fact that this mannan shows a low positive specific rotation, +14.8”, agrees with the findings of PMR study indicating the presence of large amounts of P-linkage. Acetolgsis of mannan of P. pastoris IF0 0948 strain under Conditions la and -b. In the present study, we at first attempted the reexamination of the acetolysis conditions using the P. pa&o& IF0 0948 strain mannan in accordance with the description by Gorin et al. (8) who described the acetolysis of the P. pastoris PRL 63-208 strain mannan by means of a 10/10/l (v/v) acetolysis medium at 25°C. As expected, acetolysis of
specimen
the former mannan under Condition Ia using a 10/10/l (v/v) acetolysis medium at 40°C gave only trace amounts of phosphate-containing oligosaccharide(s) eluted in the void-volume region upon gel chromatography on a column of Toyopearl HW-40 (Fig. 2A), giving rise to mannose, mannobiose, and mannopentaose as the
i
:
I 1, LI
I
II
5.5
I
I
I
PPM
I
5.0
I
I
III
4.5
FIG. 1. ‘H signal of proton magnetic resonance spectrum of P. pastoris IF0 0948 strain mannan. This was recorded using a JEOL JNM-FX 100 spectrometer in Da0 solution at 70°C in accordance with the description by Gorin et al. (8).
498
KOBAYASHI,
SHIBATA,
FIG. 2. Elution patterns of the oligosaccharide mixtures obtained from P. pastor& IF0 0948 strain mannan by acetolysis with a 10/10/l (v/v) mixture of (CH&O)aO, CHICOOH, and HaSO, at 40°C for 13 h (A) and at 25°C for 120h (B), on a column of Toyopearl HW-40 (superfine), 2.5 X 100 cm. Mg, M1, Me, Ma, and M indicate mannopentaose, mannotetraose, mannotriose, mannobiose, and mannose, respectively. V, refers to void volume.
AND SUZUKI
servations an attempt was made to adopt milder conditions, designated Conditions II and III, to acetolyze the P. pa&n-is IF0 0948 strain mannan. Acetolgsis of mannan of P. pastoris IF0 0948 strain under Conditions IIa, -b, and -c. In a previous paper from our laboratory (19), it was shown that O-acetyl derivative of the mannan of S. cerevisiae wild-type strain (bakers’ yeast) underwent the preferential cleavage of (u-1,6 linkage by the action of an acetolysis medium of low HzS04 concentration, 50/50/l (v/v) ratio of (CH&O)zO, CH&OOH, and HzS04, at 40°C for 12 h. Therefore, we adopted the above acetolytic conditions to the P. pustwis IF0 0948 strain mannan. Acetolysis fingerprints obtained under the three conditions, IIa, -b, and -c, involving different reaction periods-3, 6, and 15 h, respectively-indicated that the amount of mannotetraose produced in each acetolysate increased significantly, although the ratio of the amount of mannotetraose to that of mannopentaose decreased upon elongation of the reaction period from 3 to 15 h (Fig. 3). Therefore, it was reasonable to judge that use of the 50/50/l (v/v) acetolysis medium was still too drastic to avoid the degradation of the mannotetraose, which was assumed to involve a more susceptible structure to acetolysis than that contained in the mannopentaose. Acetolysis of mnnan of P. postoris IF0 0948 strain under Conditions IIIa, -b, and -c. Elution profiles of the acetolysates of the O-acetylated mannan of P. pastoris IF0 0948 strain obtained by adopting Conditions IIIa, -b, and -c were quite different;
common major products; the amounts of mannotriose and mannotetraose were quite small. It was also assumed that each oligosaccharide eluted in the diffusible region consists predominantly of a-1,2 linkage, because of the lesser susceptibility of this linkage under these acetolysis conditions. Unexpectedly, the elution profile of the acetolysate obtained by Condition lb (Fig. 2B) was similar to that obtained by Condition Ia, as molar ratios of the resultant oligosaccharides under the two conditions are shown in Table II. It is evident that the amounts of mannose and mannobiose are significantly large, while that of mannotetraose is considerably small. Because the TABLE II above results were consistent with the finding by Gorin et al. (8) on yields of the MOLARRATIOSOFD-MANNO-OLIGOSACCHARIDESAND F! pa&n-is oligosaccharides from the same mannan, D-MANNOSEPRODUCEDFROMD-MANNANOF it seemed reasonable to assume that Con- IF00948 STRAINBY ACETOLYSISWITHA10/10/l (v/v) MIXTUREOF(CH&O)~O, CH,COOH, AND HaSOl AT dition Ib might be as drastic as Condition 40°C FOR13hANDAT25OCFOR120h Ia for the purpose of the selective cleavage of a-1,6 linkages leaving @-1,2-linked oliCondition MS’ M1 MS Ma M gosaccharides, and that some quite labile linkages in the resultant O-acetylated oli- 4O"C,13 h 1.00 0.37 0.98 4.53 5.62 gosaccharides undergo a partial cleavage 25"C,120 h 1.00 0.34 1.14 6.12 7.65 giving rise to 0-acetyl derivatives of lower oligosaccharides. On the basis of these oba The molar ratios are expressed with Mb as unity.
ACETOLYSIS
OF B-1,2 LINKAGE-CONTAINING
i.e., the amounts of mannose and mannobiose formed under Condition IIIb were apparently smaller than those of Conditions Ia and -b (Fig. 4). Although the existence of a considerable amount of undegraded mannan was observed in the voidvolume region of the elution profile of acetolysate of Condition IIIa, a significant increase of the amount of mannotetraose was observed. Upon elongation of the reaction period to 120 h (Condition IIIc, Fig. 4C), a very small amount of the undegraded material eluted in the void-volume region and large amounts of a mixture of oligosaccharides, from mannopentaose to mannose, were evident. It will also be pointed out in Fig. 4C that a considerable decrease was observed in the amount of mannotetraose with concomitant increases in mannobiose and mannose; the latter two were assumed to be the degradation products of the higher oligosaccharides. Table III summarizes the amounts of oligosaccharides produced under Conditions IIIa to -c. A difference in the molar ratios of these
0
70
SO
so 100 FRACTION
110 NVMBER
120
130
140
FIG. 3. Elution patterns of the oligosaccharide mixtures obtained from P. past& IF0 6948 strain mannan by acetolysis with a 56/50/l (v/v) mixture of (CH&O)xO, CHeCOOH, and H&JO, at 40°C for 3 h (A), 6 h (B), and 15 h (C), on a column of Toyopearl HW40 (superfine), 2.5 X 166 cm. Symbols were the same as in Fig. 2.
MANNAN
499
FIG. 4. Elution patterns of the oligosaccharide mixtures obtained from P. pu.stwis IF0 0948 strain mannan by acetolysis with a 166/190/l (v/v) mixture of (CH&O),O, CHICOOH, and HcSO, at 40°C for 12 h (A), 36 h (B), and 120 h (C), on a column of Toyopearl HW-40 (superfine), 2.5 X 100 cm. Symbols were the same as in Fig. 2.
oligosaccharides obtained by three acetolysis conditions seemed to indicate that the mannotetraose was sensitive to acetolysis, and readily underwent the degradation to mannotriose and mannobiose under either condition, Ia or -b. These observations led us to an assumption that the acetolysislabile mannotetraose might contain a type of @-linkage different from that in mannopentaose. Proton magnetic rescmance spectra of manno-oligosaccharides obtained by acetoly&s of mannan of P. pastoris IF0 0948 strain under Conditions IIIa to -c. Figure 5 depicts the PMR spectra of H-l region of the oligosaccharides obtained from the Oacetylated mannan of P. pastoris IF0 0948 strain by acetolysis under Conditions IIIa, -b, and -c. The parts per million values of chemical shifts of each mannopentaose were consistent with those observed on a mannopentaose, Manpal - 2Manpfil --+ 2Manp/31 - 2Manpal - 2Man, isolated from the acetolysate of P. pastoris PRL 63208 strain mannan by Gorin and his co-
500
KOBAYASHI,
SHIBATA, TABLE
MOLAR
AND
SUZUKI
III
RATIOS OF D-MANNO-OLIGOSACCHARIDES AND D-MANNOSE PRODUCED FROM D-MANNAN OF P. pa-stork WITH A 100/100/l (v/v) MIXTURE OF (CH&O),O, CHaCOOH, AND HzSO, AT 40°C FOR 12,36, AND 120 h
IF0 0948 STRAIN BY ACETOLYSIS
Condition 4O”C, 12 h 40°C 36 h 40°C 120 h
MS0
M4
M3
MZ
M
1.00
2.11
2.23 2.19
0.54 0.56 0.61
2.09 2.38 3.41
0.74
1.00 1.00
1.01
1.75
a The molar ratios are expressed with MS as unity.
workers (8), when the values given by these workers were corrected in accordance with the description by Cohen and Ballou (24) (Table IV). Mannotetraoses isolated under three different acetolysis conditions gave MS
FIG. 5. ‘H signals of proton magnetic resonance spectra of the oligosaccharides obtained from the P. pa&n-is IF0 0948 strain mannan by acetolysis with a 109/199/l (v/v) mixture of (CH&O)aO, CH.&OOH, and HaSO, at 40°C for 12 h (A), 36 h (B), and 120 h (C). Each oligosaccharide corresponding to the peaks in Fig. 4 was rechromatographed on the same Toyopearl HW-40 column to remove contaminated lower and/ or higher oligosaccharides, and the PMR spectrum was determined by means of JEOL JNM-FX 100 spectrometer in DrO solution at ?‘O’C. Symbols were the same as in Fig. 2.
identical PMR spectra, and the chemical shifts indicated that the chemical structure of this mannotetraose was Manp@l 2Manpfll - 2Manpal - ZMan, which was identical to that of a mannotetraose prepared by Gorin et ~2. (8) from Manpal 2Manp/31- BManp@l - 2Manpal - 2Man by Smith degradation as the corresponding reduced form with NaBH4. It was also evident that mannotriose was a mixture of Manp@l 2Manpal 2Man and Manpal - 2Manpal - 2Man in approximately a l/l molar ratio, and mannobiose consisted solely of Manpal - 2Man. The presence of a chemical shift at 4.90 ppm, assigned as a-1,6-linked mannose residue or fi-anomeric proton of reducing terminal, was observed in the PMR spectra of mannotrioses obtained by acetolysing the Oacetylated mannan under Conditions IIIa, -b, and -c. This finding suggests a possibility that a small amount of isomer containing fragment of a-1,6-linked backbone moiety exists in each mannotriose fraction. Methylation analysis of manno-oligosaccharides obtained from acetolysate of mannan of P. pastoris IF0 0948 strain. To provide evidence that these manno-oligosaccharides consist predominantly of 1,2linkage, a series of methylation studies was conducted in accordance with the descriptions by Hakomori (20) and Lindberg (21). As will be pointed out in Table V, all oligosaccharides were found to consist of 1,2linked mannose residues. Namely, the peak corresponding to tri-O-methyl-tri-O-acetyl mannitol occurring in each manno-oligosaccharide was the 3,4,6-tri-o-methyl1,2,5-tri-0-acetyl derivative. Peaks corresponding to 2,4,6-tri-0-methyl-1,3,5-
ACETOLYSIS
OF
@-1,2
LINKAGE-CONTAINING TABLE
ANOMERIC Sugar E
D
ManpcJ
-
CHEMICAL
IV
SHIFTS FOR MANNO-OLIGOSACCHARIDES
residue
Manpfll 2Manp/31-
6 (ppm) C
M8 M8 M4 M6
PROTON
-
Manpal Manpfll 2Manpfll-t 2Manpj31-
B
-
A
Manpal 2ManpcYl 2Manpcxl 2Manpcul 2Manpcvl
+ -
2Man 2Man &Man 2Man 2Man
E
5.36
D
C
B
Aa
Afl
4.82 4.87
5.05 4.76 4.82 4.82
5.04 5.27 5.15 5.13 5.14
5.34 5.34 5.34 5.32 5.36
4.88 4.90 4.96 -
rides in a 10/10/l (v/v) mixture of (CH&0)20, CH&OOH, and H2S04. As depicted in Fig. 6, mannotetraose disappeared at 12 h, while the degradation ratio of mannopentaose was approximately 70% at the same reaction period. Therefore, this finding is consistent with the result given in Fig. 2; i.e., only a small amount of mannotetraose existed in the acetolysates of this mannan under Conditions Ia and -b. Furthermore, decrease of the amount of mannopentaose was also evident to give large amounts of mannobiose and mannose under Conditions Ia and -b and involving a prolonged reaction period.
tri-0-acetyl, 2,3,4-tri-O-methyl-1,5,6-tri-Oacetyl,and3,4-di-0-methyl-1,2,5,6-tetra-Oacetyl derivatives were not detected in the methylation products of all manno-oligosaccharides except mannotriose obtained under Condition IIIb, which contained trace amounts of the second former derivative. Specific rotations of these oligosaccharides summarized in Table V indicate that the number ratio of p-1,2 linkages to a-1,2 linkages in each oligosaccharide increased in the order mannobiose < mannotriose < mannopentaose < mannotetraose, which is consistent with the results of PMR study.
Determinatirm of degradation rates of mannotetraose and mannopentaose in IO/ 10/l (v/v) acetolysis medium. In order to
DISCUSSION
It was shown by Gorin et al. (8) that a methanol-assimilating yeast, P. pa&x-is PRL 63-208 strain, was able to elaborate a mannan containing both a-1,2 and p-1,2 linkages in the branching moieties. These workers further demonstrated that this mannan gave mannose, mannobiose, man-
provide evidences for higher lability of Manpfil - ZManp@l - 2Manpal - 2Man than Manpcyl - 2Manpfil - 2ManpPl 2Manpczl - 2Man during acetolysis, a comparative study was conducted on the difference of degradation rates between the 0-acetyl derivatives of two oligosacchaTABLE METH~LATION
Partially mannitol
acetylated derivative
2,3,4,6-Tetra-O-methyl 2,4,6-Trio-methyl 3,4,6-Tri-O-methyl 2,3,4-Tri-@methyl
[a]$ (“1
501
MANNAN
V
ANALYSIS AND SPECIFIC ROTATION OF ACETOLYSIS OLIGOSACCHARIDES CONDITION IIIb FROM P. past& IF0 0948 STRAIN MANNAN Relative retention time 1.00 1.66 1.88 2.21
W!
M8
1.00 0.83 +42.1
OBTAINED
M4
1.00 -
-
1.87 Trace
-
+14.4
UNDER
MS
1.00
1.00 -
2.69
4.01 -
-17.2
-0.5
502
KOBAYASHI,
SHIBATA,
FIG. 6. Time course of acetolysis of mannotetraose (0) and mannopentaose (0) with a 10/10/l (v/v) mixture of (CH&O)cO, CHaCOOH, and HaSO, at 40°C. These oligosaccharides were obtained from the P. pastoris IF0 0948 strain mannan by acetolysis with a 109/199/l (v/v) mixture of (CH&O)aO, CH&OOH, and H&SO, at 40°C for 36 h. Degradation ratios of oligosaccharides were calculated with the formula described under Materials and Methods.
notriose, and mannopentaose upon partial acetolysis under the same conditions as those of Condition Ib, and that the chemical structures of mannobiose and mannopentaose were ManpcYl - 2Man and Manpal - 2Manppl 2Manp/31 2Manpal - ZMan, respectively. However, the results obtained in the present study indicated that the molar ratio of the oligosaccharides obtained by acetolysis under Condition Ia was closely identical to that of the oligosaccharides obtained under Condition Ib. Therefore, it was reasonable to consider that Condition Ib was too drastic to leave /3-1,2-linked manno-oligosaccharides as well as Condition Ia. This finding also suggested that the decrease in H&SO4 concentration in acetolysis medium might be effective for controlling the acetolytic rate. We therefore attempted an investigation to find out the proper acetolysis condition for the isolation of manno-oligosaccharides containing p-1,2 linkage without serious degradation. One of the significant results obtained in the present study was the increase in the amount of mannotetraose under Condition III which was scarcely produced under Conditions Ia and -b from the same mannan. The results of PMR, specific rotation,
AND
SUZUKI
and methylation studies provided evidence that the structures of mannotetraose and mannopentaose were Manpbl 2Manpfll - 2Manpcwl - 2Man and Manpcrl - 2ManpPl 2ManpPl 2Manpal - ZMan, respectively. It is also of interest that the difference in the susceptibility of these oligosaccharides to acetolysis is considerably large, providing evidence that the nonreducing terminal sugar residue connected to the p-1,2 linkage is more sensitive to acetolysis than that located in the intermediary position connected to the same linkage. This finding seems to agree with the findings reported by Zhang and Ballou (25) on the prompt degradation of fi-1,2-linked manno-oligosaccharides with a 10/10/l mixture of (CH&O)PO, CH&OOH, and H,SO,; i.e., the relative acetolytic rate of Manpoll - BMan, Manp@l - BMan, and Manpal - 6Man was approximately 30/2/l. Additionally, it was observed in the present study that, even under a mild acetolysis condition such as Condition III, the degradation of Manppl - 2Manppl - 2Manpcrl - 2Man took place at a measurable rate. Therefore, depiction of the chemical structure of the mannan of P. p&or& IF0 0948 strain on the basis of findings from acetolysis studies should be made by taking account of the high susceptibility of Manppl - 2ManpPl + 2Manpal - 2Man to acetolysis, even milder conditions were adopted to the acetolysis.
3 4 FIG. 7. Representative structure of the P. pasti IF0 0948 strain mannan. M denotes a D-mannopyranosyl residue. The numbers given outside the brackets indicate the approximate number of side chains, the sequences of which are not specified.
ACETOLYSIS
OF
b-1,2
LINKAGE-CONTAINING
Summarizing all findings obtained by Gorin et al. (8) and by us, a representative chemical structure of the mannan of P. pastoris IF0 0948 strain can be depicted as shown in Fig. 7. It is obvious that the longest side chain in the P. pastoris IF0 0948 strain mannan was Manwl - 2Manpal 2Manpfil - 2Manpcyl - BMan, although the predominant oligosaccharide detected in the acetolysate was Manp/31 2ManpPl - 2Manpal - BMan, which was found to be about 2 times larger than the former in the molar ratio. Because the content of alkali-eliminable oligosaccharide residues in P. pastoris IF0 0948 strain mannan was found to be low, less than 1.0% (data not shown), it is reasonable to state that these oligosaccharide residues can be disregarded as a depiction of the chemical structure of the mannan moiety. Conclusively, the use of acetolysis medium containing H2S04 at a relatively low concentration, up to 0.5% (v/v), in which occurs the preferential cleavage of the a1,6 linkage in a yeast mannan giving rise to manno-oligosaccharides containing acetolysis-labile /l-1,2 linkage seems to be advantageous for the structural investigation of mannose-containing polysaccharides and high-mannose-type carbohydrate moieties of many glycoproteins. REFERENCES 1. SUMMERS, D. F., GROLLMAN, A. P., AND HASENCLEVER, H. F. (1964) J. Immu?wL 92,491-499. 2. HASENCLEVER, H. F., AND MITCHELL, W. 0. (1964)
J. Immwwl
93,763-771.
3. LEE, Y., AND BALLOU, C. E. (1965) Biochemistry 4, 257-264. 4. SUZUKI, S., SUNAYAMA, H., AND SAITO, S. (1968) Japan J. Microbial 12,19-X 5. KOCOUREK, J., AND BALLOU, C. E. (1969) J. BacterioL 100,11’75-1181.
503
MANNAN
6. GORIN, P. A. J., AND PERLIN, A. S. (1956) Can&. J. 0x-m 34,1796-1803. 7. SUZUKI, S., AND SUNAYAMA, H. (1968) Japan. J. Microbial 12,413-422. 8. GORIN, P. A. J., SPENCER, J. F. T., AND BHATTACHARJEE, S. S. (1969) Canad. J. Cfiem 47,14991505. 9. HAMADA, T., NAKAJIMA, T., IZAKI, K., AND MATSUDA, K. (1981) Eur. J. B&hem 119,365-371. 10. FUNAYAMA, M., NISHIKAWA, A., SHINODA, T., SuZUKI, M., AND FUKAZAWA, Y. (1984) Micro&o1 Immunol 28,1359-13’71. 11. GORIN, P. A. J., AND SPENCER, J. F. T. (1970) Adw. Appl. Micr&iol 13,25-89. 12. DUBOIS, M., GIL.LES, K. A., HAMILTON, J. K., REBERS, P. A., AND SMITH, F. (1956) Anal Chem 28,350-356. 13. AMES, B. N., AND DUBIN, D. T. (1960) J. Biol. Chem 235,769-775. 14. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem 193, 265-275. 15. WEILL, C. E., AND HANKE, P. (1962) Anal. Chem 34,1736-1737. 16. OKUBO, Y., SHIBATA, N., ICHIKAWA, T., CHAKI, S., AND SUZUKI, S. (1981) Arch B&hem. Biophys. 212,204-215. 17. ISHITANI, T., AND HIRATA, T., AND KATO, K. (1978)
Agri. Biol. Chem. 42,897-898. 18. OKUBO,
Y., ICHIKAWA,
T., AND SUZUKI,
S. (1978)
J.
Bacterial. 136, 63-68. 19. OKUBO, Y., AND SUZUKI, S. (1978) 62,135-141. 20. HAKOMORI, S. (1946) J. B&hem. 208.
Carbohydr. Res. (Tokyo) 55,205-
21. LINDBERG, B. (1972) in Methods in Enzymology (Ginsburg, V., ed.), Vol. 28, Part B, pp. 178-195, Academic Press, New York/London. 22. STEWART, T. S., AND BALLOU, C. E. (1968) Biochemistry 7,1855-1863. 23. JONES, G. H., AND BALLOU, C. E. (1969) J. Biol
Chem. 244,1052-1059. 24. COHEN,
R. E., AND BALLOU,
C. E. (1980) Biochem-
istry 19,4345-4358. 25. ZHANG, W., AND Bwou, 256,10073-10079.
C. E. (1981)
J.
Bid Chem
Acetolysis of Pichia pastoris IF0 0948 Strain Mannan Containing a-l,2 and ,f3-1,2 Linkages Using Acetolysis Medium of Low Sulfuric Acid Concentration HIDEMITSU second Department
KOBAYASHI, of Hygienic
NOBUYUKI SHIBATA,
AND
Ch.emist?y, Tohoku College of Pharmacy,
SHIGEO SUZUKI’ S&i,
Miyagi
983, Japan
Received June 21, 1985, and in revised form October 22,1985
To obtain manno-oligosaccharides containing @-1,2-linked nonreducing terminal groups from the mannan of Pichia pastoris IF0 0948 strain by acetolysis, an attempt was made to establish the reaction conditions under which cleavage of the a-1,6 linkage took place preferentially leaving manno-oligosaccharides composed largely of p-1,2 linkages. By the action of an ordinary acetolysis medium, a 10/10/l (v/v) mixture of acetic anhydride, acetic acid, and sulfuric acid at 40°C for 13 h or at 25°C for 120 h, the Oacetyl derivative of this mannan gave mannose, mannobiose, mannotriose, and mannopentaose. However, treatment of the same O-acetyl mannan with a 50/50/l (v/v) acetolysis medium at 40°C for 15 h gave a mannotetraose in addition to mannose, mannobiose, mannotriose, and mannopentaose. Use of a 100/100/l (v/v) acetolysis medium at 40°C for 36 h gave a more satisfactory result, a mixture of oligosaccharides, from mannose to mannopentaose, which contained more mannotetraose than mannopentaose. Because both mannotetraose and mannopentaose contained a-1,2 and p-1,2 linkages, it was concluded that an acetolysis medium containing a low concentration of sulfuric acid, up to 0.5% (v/v), facilitates the preferential cleavage of the a-1,6 linkage, leaving manno-oligosaccharides containing the B-1,2 linkage which was found to be labile to the action of the 10/10/l (v/v) acetolysis medium. o 19s Academic PW., IN.
It is well known that cell wall mannans comprise the principal antigens of the parent yeast cells (1, 2), and that controlled acetolysis provides a convenient method for the analysis of immunochemically specific sites corresponding to the branching moieties of the parent mannans (3,4) and for determining the chemical structure of the intact mannans (5). Gorin and Perlin (6) first applied acetolysis to the structural study of yeast mannans, and were able to isolate Manpal - 2Man.’ Thereafter, 1 To whom all correspondence should be addressed. ’ Abbreviations used: PMR, proton magnetic resonance; Manpal + 2Manppl 2Manpfil 2Manpal - 2Man, ~~-D-mSnnOpyranOSyl-(l 2)~~-D-mannOpyrSnOSyl-(l
nosyl-(1
-
-
2)-~~-D-mannOpyra-
2)-C&x-D-mannopyranosyl-(1
0003-9861/86 $3.00 Copyright All righta
Q 1986 by Academic Press, Inc. of reproduction in any form reserved.
-
2)-D-
many papers were published using acetolysis to study various yeast mannans (4,5, 7-10). Some yeast mannans, however, were shown to contain @linked mannose residues from the results of proton magnetic resonance (PMR) studies by Gorin and Spencer (11). Structural information obmannose; Man@1 - 2Manp@l2Manpal2Man, O-B-D-mannopyranosyl-(1 + 2)-@@-D-mannopyra- 2)-Dnosyl-(1 - 2)-&x-D-mannopyranosyl-(1 mannose; Man@1 - 2Manpal - BMan, o-b-D-mannopyranosyl-(1 - 2)-&x-D-mannopyranosyl-(l 2)D-mannose; Manpal - 2Manpal - BMan, &Y-Dmannopyranosyl-(1 - 2)-Oa-D-mannopyranosyl-(l + 2)-D-IIUmnOSe; Manpal - SMan, O-a-D-mannopyranosyl-(1 - 2)-D-mannose; Manp~l - BMan, o-@-Dmannopyranosyl-(1 - 2)-D-mannose; Manpal -+ GMan, ~cY-D-mSnnOpyrSnOSyl-(1-+ 6)-D-mannose. 494
ACETOLYSIS
OF b-1,2 LINKAGE-CONTAINING
MANNAN
495
tained by acetolysis on these mannans was was made by means of a Shimadzu Chromatopac-ElA limited in another report by Gorin and his microcomputer. Materials. The P. pastoris IF0 0948 strain was obco-workers (8). The results of our investitained from Institute for Fermentation, Osaka, Japan. gation on acetolysis conditions for the A column packing for gel-filtration chromatography, mannan from Pichia pastwis IF0 0948 Toyopearl HW-40 (superfine), fractionation range of strain, which show a PMR pattern closely 100 to 7000 Da, was purchased from Toyo Soda Manidentical to that of the P. pastoris PRL 63- ufacturing Company, Ltd., Tokyo, Japan. 208 strain mannan published by Gorin et Cultivation of P. pastoris IF0 0948 strain The P. al (8), showed discrepancies with those ob- pastoris IF0 0948 strain was grown in 500-ml flasks tained by these workers (8). For example, containing a liquid medium described by Ishitani et the chemical structure of this mannan as al (17) on a reciprocal shaker at 28°C. The composition proposed by these workers (8) contains only of the liquid medium was as follows: NH&l, 4.0 g; one kind of branching moiety correspond- KaHPO,, 2.0 g; KHaPO,, 2.0 g; MgSOd.?HaO, 0.5 g; corn steep liquor (Sigma Chemical Co., St. Louis, MO.), ing to a pentaose, Manpal - 2ManpPl 2.0 g; glucose, 20 g; in 1000 ml of water (pH 6.0). The 2Manp/31 - 2Manpal - BMan, although cells were harvested after 48 h, washed three times many yeast mannans were found to contain with saline by centrifugation, and dehydrated with a branching moieties of different length (5). large volume of acetone. In this paper, we describe a modified acPreparation of the mannan of P. pastor6 IF0 09.48 etolysis method with milder conditions for strain This was completed in accordance with the previous description for the isolation of a X cerevisiae the isolation of manno-oligosaccharides wild-type strain (18) by means of Fehling solution. composed predominantly of fi-1,2-linked Yield of the mannan was 12.5% on weight basis of the mannose residues including Manpal 2Manp@l - 2Manpcvl - 2Man from the acetone-dried whole cells. Acetolysis of mannan, Prior to acetolysis, the manmannan of P. pastoris IF0 0948 strain. nan was converted into its 0-acetyl derivative in acMATERIALS
AND
METHODS
General. Total carbohydrate and total phosphate were determined by the phenol-sulfuric acid method (12) and by the Ames-Dubin method (13), respectively. Total protein was determined by the Folin method of Lowry et al (14) using bovine serum albumin as the standard. Precoated silica gel plates (Kiesel Gel, without fluorescent indicator, 0.25-mm thickness, 20 X 20 cm, Merck, Darmstadt, West Germany) were used for thin-layer chromatography. The solvent used for thin-layer chromatography was a 5/3/2 (v/v) mixture of 1-butanol-ethanol-water. Sugars on the plates were detected by spraying with anisaldehyde reagent (15). Specific rotations were measured using an Applied Electric automatic polarimeter (c = 1.0, 1= 1.0, water). PMR spectra of the H-l region of mannan and mannooligosaccharides were recorded using a JEOL JNMFX 100 spectrometer in DsO solution at 70°C in accordance with the description by Gorin and Spencer (11). Sugar component analysis of the mannan of P. pastoris IF0 0948 strain was conducted in accordance with the description by Okubo et al (16) on that of the mannan of a Saccharomyces cerevi.siae wild type strain prepared by means of cetyltrimethylammonium bromide. Gas chromatography of Omethyl-0-acetyl mannitols was conducted on a glass column (5 mm X 150 cm) containing 3% OV-210 on Supelcoport (100 to 200 mesh) at 185°C using N2 as the carrier gas at a flow rate of 20 ml/min. Conversion of the peak dimensions into molar ratios of the sugar derivatives
cordance with the description by Okubo and Suzuki (19). Namely, mannan, 100 mg, and anhydrous formamide, 5 ml, contained in a 50-ml round-bottomed flask with a glass stopper were heated in a boiling water bath until the solid mass was dissolved, and the resultant solution was cooled to room temperature. To the solution was added a l/l (v/v) mixture of (CH&O)aO and anhydrous pyridine, 20 ml, at once, and the clear solution was kept at 40°C for 12 h. Then the reaction mixture was concentrated in nacho to a thick syrup, and the rest of the volatile materials was thoroughly removed by means of an oil diffusion pump to leave the 0-acetylated mannan as a hard glass in quantitative yield. Then the Oacetylated mannan was subjected to acetolysis under eight different conditions, designated Conditions Ia, Ib, IIa, IIb, IIc, IIIa, IIIb, and 111~ as follows: In Condition Ia, the 0-acetyl mannan was dissolved in a 10/10/l (v/v) mixture of (CH&O),O, CHaCOOH, and HaSO,, 10 ml, and the resultant solution was kept at 40°C for 13 h. This condition was the same as that described by Kocourek and Ballou (5). After termination of the reaction by the addition of pyridine, 2 ml, the mixture was concentrated in vucuo to a thick syrup, and dissolved in CHCls, 10 ml. Then the solution was shaken with water, each 15 ml, three times, and dried over anhydrous NaaSO,. After filtration, the solvent was evaporated in vocv.~ to dryness; then trace of CHCls was removed by codistillation with toluene, 5 ml. The residue was dissolved in anhydrous CHaOH, 2.5 ml, and 1 M methanol solution of CH80Na was added dropwise until the precipitation of de-O-acet-
496
KOBAYASHI,
SHIBATA,
ylation product started. After leaving for 30 min, the mixture was neutralized with 50% CH&OOH, and evaporated in vacua to dryness to yield a mixture of manno-oligosaccharides containing small amounts of CH&OONa. In Condition Ib, the composition of the acetolysis medium was the same as that of Condition Ia, while the reaction temperature and period were 25°C and 120 h, respectively. This condition was exactly identical to that for the structural study of the mannan of P. pastoris PRL 63-208 strain reported by Gorin et al. (8). Conditions IIa, -b, and -c involved treatment of the 0-acetyl mannan with a 50/50/l (v/v) mixture of (CH&O)zO, CI-I&OOH, and H&SO4 at 40°C for three different periods-3,6, and 15 h, respectively-in accordance with the description by Okubo and Suzuki (19) on a controlled acetolysis study of the mannan of a S. cerevisioe wild-type strain. Conditions IIIa, -b, and -c involved the use of an acetolysis medium of lower HzSOl concentration, a 109/100/l (v/v) mixture of (CI-I&O)zO, CH&OOH, and HzSO, at 40°C for three different reaction periods, 12, 36, and 120 h, which were designated Conditions IIIa, -b, and -c, respectively. The de-0-acetylated oligosaccharide mixture was dissolved in water, 2.0 ml, and was applied to a column of Toyopearl HW-40 (2.5 X 100 cm). Elution was made with water (0.25 ml/min), and aliquots of the eluates, each 15 ~1, were assayed for carbohydrate content by the phenol-sulfuric acid method (12). Eluates corresponding to each peak were combined, concentrated in w~1cuo to a small volume, and assayed for the homogeneity of the oligosaccharides contained in the solution by thin-layer chromatography. To remove small amounts of contaminated oligosaccharides of lower or higher molecular weight, the solution was rechromatographed on the same column of Toyopearl HW-40, and the eluates containing a homogeneous oligosaccharide were combined and lyophilized after concentration in vacua. Meth&&m an&s&. Methylation of manno-oligosaccharides, each 5 mg, obtained by gel filtration of the acetolysis products was conducted by the Hakomori method (20), and the resultant per-O-methylated manno-oligosaccharides were converted into a mixture of 0-methyl-0-acetyl mannitols according to the description by Lindberg (21).
AND
SUZUKI
resultant solution was kept at 40% At intervals of 1, 3, 6, 12, and 24 h, aliquots of solution, each 5 ml, were withdrawn, and were evaporated in vacua after addition of pyridine, 1.0 ml. The thick syrup was dissolved in CHCl,, 10 ml, and washed with water, 100 ml. The CHCh, layer was separated and dried over anhydrous Na#O,. After filtration, the solution was evaporated in vacua to dryness, and the residue was dissolved in anhydrous CHzOH, 3 ml. To the solution was added a few drops of 1 M methanolic CHsONa solution, and the mixture was left at room temperature until the precipitation of free sugars was accomplished. Then the mixture was neutralized with 50% CHaCOOH, and evaporated in v-0 to dryness. The residue was dissolved in water, 10 ml, and was deionized by the addition of a mixture of Amberlite IR-120 (H+) and Amberlite IR-410 (OH-) resins. The supernatant was collected by filtration, and the resin mixture was thoroughly washed with water. Then the combined filtrate and washing were concentrated in vaouo to dryness. After dissolving in a 68132 (v/v) mixture of CHsCN and water, 200 ~1, the solution was analyzed by high-performance liquid chromatography using an Erma ERC-8710 liquid chromatograph equipped with a column of ERC-NH-11’71,6 X 200 mm, and a differential refractometer. Elution was made with the same mixture of CHaCN and water at 100 kg/cm’, and the eluate wae monitored for the amounts of unchanged and degraded oligosaccharides. Under the chromatographic conditions, the relative retention times of mannose, mannobiose, mannotriose, mannotetraose, and mannopentaose were 1.00, 1.16, 1.48, 1.81, and 2.06, respectively. Degradation ratios of the starting mannotetraose or mannopentaose at five interval periods were computed by the following formula: Degradation ratio (a)= A/(A + B) X 100, where A represents sum of the peak dimensions of the degradation products including lower oligosaccharides than each corresponding parent oligosaccharide and mannose, and B indicates the peak dimension of unchanged parent oligosaccharide. It should be noted that a 100% degradation means the disappearance of each starting manno-oligosaccharide, tetraose, or pentaose in the reaction mixture, and not complete degradation to mannose. RESULTS
Determination of acetolytic rates of mmnotetraose and mannqpentaose in aatolysis medium of a lO/lO/ 1 (v/v) mixture of @H&O)&, CHJOOH, and HSO,.
Properties of the mannan of P. passtoris IF0 0948 strain. The results of chemical
Each oligosaccharide, 20 mg, was dissolved in 5 ml of anhydrous formamide in a 50-ml glass-stoppered, round-bottomed flask. To the solution was added a 11 1 (v/v) mixture of (CH&O)zO and anhydrous pyridine, 10 ml, on a boiling water bath until the solid mass dissolved. Then the clear solution was evaporated in vacua to dryness using an oil diffusion pump. The residue was dissolved in a mixture of (CH&O)zO, CH&OOH, and H&SO,, 10/10/l (v/v) ratio, and the
analyses of the mannan of P. pastoris IF0 0948 strain are shown in Table I. The sugar component analysis of the mannan revealed that mannose was the major sugar; the presence of a small amount of glucose was also evident. The mannan contained more than 90% total carbohydrate, 3.0% protein, and trace amounts of phosphate.
ACETOLYSIS
OF o-1,2 LINKAGE-CONTAINING TABLE
497
MANNAN
I
CHEMICAL COMPOSITIONOF THE MANNAN OBTAINED FROM P. pastoris
Total carbohydrate
Molar ratio of sugar eomponentb
(%I”
Mannose
Glucose
93.0
97.0
3.0
Total protein (%I”
Total phosphate (%ld
l@
Trace
f14.8
3.0
(“)”
a Determined by the phenol-sulfuric acid method (12). b Determined by gas chromatography by converting the component sugars of each acid-hydrolyzed into the corresponding aiditol acetates (16). ‘Determined bv the Folin method of Lowry et al. (14). d Determined by the Ames-Dubin method (13). BIn water, c, 1.0; 1, 1.0.
Therefore, this mannan was expected to give very small amounts of phosphatecontaining oligosaccharide(s) eluted in the void-volume region in the acetolysis fingerprint of its acetolysate obtained by gel chromatography as observed by a few workers on several yeast mannans (l&22, 23). This mannan also gave a PMR pattern closely identical to that of the P. pastoris PRL 63-208 strain mannan, demonstrating the existence of large amounts of (u-1,2and p-1,2 linkages corresponding to chemical shifts of 5.09 to 5.38 and 4.87 ppm in the PMR pattern of this mannan, respectively (Fig. 1). However, it should be noted that this PMR pattern is not consistent with the chemical structure of P. pastoris PRL 63208 strain proposed by Gorin et al. (8), because it was shown to contain a single kind of branching moiety corresponding to a mannopentaose residue, Manpcvl 2ManpPl - 2ManpBl - 2Manpal - 2Man. The fact that this mannan shows a low positive specific rotation, +14.8”, agrees with the findings of PMR study indicating the presence of large amounts of P-linkage. Acetolgsis of mannan of P. pastoris IF0 0948 strain under Conditions la and -b. In the present study, we at first attempted the reexamination of the acetolysis conditions using the P. pa&o& IF0 0948 strain mannan in accordance with the description by Gorin et al. (8) who described the acetolysis of the P. pastoris PRL 63-208 strain mannan by means of a 10/10/l (v/v) acetolysis medium at 25°C. As expected, acetolysis of
specimen
the former mannan under Condition Ia using a 10/10/l (v/v) acetolysis medium at 40°C gave only trace amounts of phosphate-containing oligosaccharide(s) eluted in the void-volume region upon gel chromatography on a column of Toyopearl HW-40 (Fig. 2A), giving rise to mannose, mannobiose, and mannopentaose as the
i
:
I 1, LI
I
II
5.5
I
I
I
PPM
I
5.0
I
I
III
4.5
FIG. 1. ‘H signal of proton magnetic resonance spectrum of P. pastoris IF0 0948 strain mannan. This was recorded using a JEOL JNM-FX 100 spectrometer in Da0 solution at 70°C in accordance with the description by Gorin et al. (8).
498
KOBAYASHI,
SHIBATA,
FIG. 2. Elution patterns of the oligosaccharide mixtures obtained from P. pastor& IF0 0948 strain mannan by acetolysis with a 10/10/l (v/v) mixture of (CH&O)aO, CHICOOH, and HaSO, at 40°C for 13 h (A) and at 25°C for 120h (B), on a column of Toyopearl HW-40 (superfine), 2.5 X 100 cm. Mg, M1, Me, Ma, and M indicate mannopentaose, mannotetraose, mannotriose, mannobiose, and mannose, respectively. V, refers to void volume.
AND SUZUKI
servations an attempt was made to adopt milder conditions, designated Conditions II and III, to acetolyze the P. pa&n-is IF0 0948 strain mannan. Acetolgsis of mannan of P. pastoris IF0 0948 strain under Conditions IIa, -b, and -c. In a previous paper from our laboratory (19), it was shown that O-acetyl derivative of the mannan of S. cerevisiae wild-type strain (bakers’ yeast) underwent the preferential cleavage of (u-1,6 linkage by the action of an acetolysis medium of low HzS04 concentration, 50/50/l (v/v) ratio of (CH&O)zO, CH&OOH, and HzS04, at 40°C for 12 h. Therefore, we adopted the above acetolytic conditions to the P. pustwis IF0 0948 strain mannan. Acetolysis fingerprints obtained under the three conditions, IIa, -b, and -c, involving different reaction periods-3, 6, and 15 h, respectively-indicated that the amount of mannotetraose produced in each acetolysate increased significantly, although the ratio of the amount of mannotetraose to that of mannopentaose decreased upon elongation of the reaction period from 3 to 15 h (Fig. 3). Therefore, it was reasonable to judge that use of the 50/50/l (v/v) acetolysis medium was still too drastic to avoid the degradation of the mannotetraose, which was assumed to involve a more susceptible structure to acetolysis than that contained in the mannopentaose. Acetolysis of mnnan of P. postoris IF0 0948 strain under Conditions IIIa, -b, and -c. Elution profiles of the acetolysates of the O-acetylated mannan of P. pastoris IF0 0948 strain obtained by adopting Conditions IIIa, -b, and -c were quite different;
common major products; the amounts of mannotriose and mannotetraose were quite small. It was also assumed that each oligosaccharide eluted in the diffusible region consists predominantly of a-1,2 linkage, because of the lesser susceptibility of this linkage under these acetolysis conditions. Unexpectedly, the elution profile of the acetolysate obtained by Condition lb (Fig. 2B) was similar to that obtained by Condition Ia, as molar ratios of the resultant oligosaccharides under the two conditions are shown in Table II. It is evident that the amounts of mannose and mannobiose are significantly large, while that of mannotetraose is considerably small. Because the TABLE II above results were consistent with the finding by Gorin et al. (8) on yields of the MOLARRATIOSOFD-MANNO-OLIGOSACCHARIDESAND F! pa&n-is oligosaccharides from the same mannan, D-MANNOSEPRODUCEDFROMD-MANNANOF it seemed reasonable to assume that Con- IF00948 STRAINBY ACETOLYSISWITHA10/10/l (v/v) MIXTUREOF(CH&O)~O, CH,COOH, AND HaSOl AT dition Ib might be as drastic as Condition 40°C FOR13hANDAT25OCFOR120h Ia for the purpose of the selective cleavage of a-1,6 linkages leaving @-1,2-linked oliCondition MS’ M1 MS Ma M gosaccharides, and that some quite labile linkages in the resultant O-acetylated oli- 4O"C,13 h 1.00 0.37 0.98 4.53 5.62 gosaccharides undergo a partial cleavage 25"C,120 h 1.00 0.34 1.14 6.12 7.65 giving rise to 0-acetyl derivatives of lower oligosaccharides. On the basis of these oba The molar ratios are expressed with Mb as unity.
ACETOLYSIS
OF B-1,2 LINKAGE-CONTAINING
i.e., the amounts of mannose and mannobiose formed under Condition IIIb were apparently smaller than those of Conditions Ia and -b (Fig. 4). Although the existence of a considerable amount of undegraded mannan was observed in the voidvolume region of the elution profile of acetolysate of Condition IIIa, a significant increase of the amount of mannotetraose was observed. Upon elongation of the reaction period to 120 h (Condition IIIc, Fig. 4C), a very small amount of the undegraded material eluted in the void-volume region and large amounts of a mixture of oligosaccharides, from mannopentaose to mannose, were evident. It will also be pointed out in Fig. 4C that a considerable decrease was observed in the amount of mannotetraose with concomitant increases in mannobiose and mannose; the latter two were assumed to be the degradation products of the higher oligosaccharides. Table III summarizes the amounts of oligosaccharides produced under Conditions IIIa to -c. A difference in the molar ratios of these
0
70
SO
so 100 FRACTION
110 NVMBER
120
130
140
FIG. 3. Elution patterns of the oligosaccharide mixtures obtained from P. past& IF0 6948 strain mannan by acetolysis with a 56/50/l (v/v) mixture of (CH&O)xO, CHeCOOH, and H&JO, at 40°C for 3 h (A), 6 h (B), and 15 h (C), on a column of Toyopearl HW40 (superfine), 2.5 X 166 cm. Symbols were the same as in Fig. 2.
MANNAN
499
FIG. 4. Elution patterns of the oligosaccharide mixtures obtained from P. pu.stwis IF0 0948 strain mannan by acetolysis with a 166/190/l (v/v) mixture of (CH&O),O, CHICOOH, and HcSO, at 40°C for 12 h (A), 36 h (B), and 120 h (C), on a column of Toyopearl HW-40 (superfine), 2.5 X 100 cm. Symbols were the same as in Fig. 2.
oligosaccharides obtained by three acetolysis conditions seemed to indicate that the mannotetraose was sensitive to acetolysis, and readily underwent the degradation to mannotriose and mannobiose under either condition, Ia or -b. These observations led us to an assumption that the acetolysislabile mannotetraose might contain a type of @-linkage different from that in mannopentaose. Proton magnetic rescmance spectra of manno-oligosaccharides obtained by acetoly&s of mannan of P. pastoris IF0 0948 strain under Conditions IIIa to -c. Figure 5 depicts the PMR spectra of H-l region of the oligosaccharides obtained from the Oacetylated mannan of P. pastoris IF0 0948 strain by acetolysis under Conditions IIIa, -b, and -c. The parts per million values of chemical shifts of each mannopentaose were consistent with those observed on a mannopentaose, Manpal - 2Manpfil --+ 2Manp/31 - 2Manpal - 2Man, isolated from the acetolysate of P. pastoris PRL 63208 strain mannan by Gorin and his co-
500
KOBAYASHI,
SHIBATA, TABLE
MOLAR
AND
SUZUKI
III
RATIOS OF D-MANNO-OLIGOSACCHARIDES AND D-MANNOSE PRODUCED FROM D-MANNAN OF P. pa-stork WITH A 100/100/l (v/v) MIXTURE OF (CH&O),O, CHaCOOH, AND HzSO, AT 40°C FOR 12,36, AND 120 h
IF0 0948 STRAIN BY ACETOLYSIS
Condition 4O”C, 12 h 40°C 36 h 40°C 120 h
MS0
M4
M3
MZ
M
1.00
2.11
2.23 2.19
0.54 0.56 0.61
2.09 2.38 3.41
0.74
1.00 1.00
1.01
1.75
a The molar ratios are expressed with MS as unity.
workers (8), when the values given by these workers were corrected in accordance with the description by Cohen and Ballou (24) (Table IV). Mannotetraoses isolated under three different acetolysis conditions gave MS
FIG. 5. ‘H signals of proton magnetic resonance spectra of the oligosaccharides obtained from the P. pa&n-is IF0 0948 strain mannan by acetolysis with a 109/199/l (v/v) mixture of (CH&O)aO, CH.&OOH, and HaSO, at 40°C for 12 h (A), 36 h (B), and 120 h (C). Each oligosaccharide corresponding to the peaks in Fig. 4 was rechromatographed on the same Toyopearl HW-40 column to remove contaminated lower and/ or higher oligosaccharides, and the PMR spectrum was determined by means of JEOL JNM-FX 100 spectrometer in DrO solution at ?‘O’C. Symbols were the same as in Fig. 2.
identical PMR spectra, and the chemical shifts indicated that the chemical structure of this mannotetraose was Manp@l 2Manpfll - 2Manpal - ZMan, which was identical to that of a mannotetraose prepared by Gorin et ~2. (8) from Manpal 2Manp/31- BManp@l - 2Manpal - 2Man by Smith degradation as the corresponding reduced form with NaBH4. It was also evident that mannotriose was a mixture of Manp@l 2Manpal 2Man and Manpal - 2Manpal - 2Man in approximately a l/l molar ratio, and mannobiose consisted solely of Manpal - 2Man. The presence of a chemical shift at 4.90 ppm, assigned as a-1,6-linked mannose residue or fi-anomeric proton of reducing terminal, was observed in the PMR spectra of mannotrioses obtained by acetolysing the Oacetylated mannan under Conditions IIIa, -b, and -c. This finding suggests a possibility that a small amount of isomer containing fragment of a-1,6-linked backbone moiety exists in each mannotriose fraction. Methylation analysis of manno-oligosaccharides obtained from acetolysate of mannan of P. pastoris IF0 0948 strain. To provide evidence that these manno-oligosaccharides consist predominantly of 1,2linkage, a series of methylation studies was conducted in accordance with the descriptions by Hakomori (20) and Lindberg (21). As will be pointed out in Table V, all oligosaccharides were found to consist of 1,2linked mannose residues. Namely, the peak corresponding to tri-O-methyl-tri-O-acetyl mannitol occurring in each manno-oligosaccharide was the 3,4,6-tri-o-methyl1,2,5-tri-0-acetyl derivative. Peaks corresponding to 2,4,6-tri-0-methyl-1,3,5-
ACETOLYSIS
OF
@-1,2
LINKAGE-CONTAINING TABLE
ANOMERIC Sugar E
D
ManpcJ
-
CHEMICAL
IV
SHIFTS FOR MANNO-OLIGOSACCHARIDES
residue
Manpfll 2Manp/31-
6 (ppm) C
M8 M8 M4 M6
PROTON
-
Manpal Manpfll 2Manpfll-t 2Manpj31-
B
-
A
Manpal 2ManpcYl 2Manpcxl 2Manpcul 2Manpcvl
+ -
2Man 2Man &Man 2Man 2Man
E
5.36
D
C
B
Aa
Afl
4.82 4.87
5.05 4.76 4.82 4.82
5.04 5.27 5.15 5.13 5.14
5.34 5.34 5.34 5.32 5.36
4.88 4.90 4.96 -
rides in a 10/10/l (v/v) mixture of (CH&0)20, CH&OOH, and H2S04. As depicted in Fig. 6, mannotetraose disappeared at 12 h, while the degradation ratio of mannopentaose was approximately 70% at the same reaction period. Therefore, this finding is consistent with the result given in Fig. 2; i.e., only a small amount of mannotetraose existed in the acetolysates of this mannan under Conditions Ia and -b. Furthermore, decrease of the amount of mannopentaose was also evident to give large amounts of mannobiose and mannose under Conditions Ia and -b and involving a prolonged reaction period.
tri-0-acetyl, 2,3,4-tri-O-methyl-1,5,6-tri-Oacetyl,and3,4-di-0-methyl-1,2,5,6-tetra-Oacetyl derivatives were not detected in the methylation products of all manno-oligosaccharides except mannotriose obtained under Condition IIIb, which contained trace amounts of the second former derivative. Specific rotations of these oligosaccharides summarized in Table V indicate that the number ratio of p-1,2 linkages to a-1,2 linkages in each oligosaccharide increased in the order mannobiose < mannotriose < mannopentaose < mannotetraose, which is consistent with the results of PMR study.
Determinatirm of degradation rates of mannotetraose and mannopentaose in IO/ 10/l (v/v) acetolysis medium. In order to
DISCUSSION
It was shown by Gorin et al. (8) that a methanol-assimilating yeast, P. pa&x-is PRL 63-208 strain, was able to elaborate a mannan containing both a-1,2 and p-1,2 linkages in the branching moieties. These workers further demonstrated that this mannan gave mannose, mannobiose, man-
provide evidences for higher lability of Manpfil - ZManp@l - 2Manpal - 2Man than Manpcyl - 2Manpfil - 2ManpPl 2Manpczl - 2Man during acetolysis, a comparative study was conducted on the difference of degradation rates between the 0-acetyl derivatives of two oligosacchaTABLE METH~LATION
Partially mannitol
acetylated derivative
2,3,4,6-Tetra-O-methyl 2,4,6-Trio-methyl 3,4,6-Tri-O-methyl 2,3,4-Tri-@methyl
[a]$ (“1
501
MANNAN
V
ANALYSIS AND SPECIFIC ROTATION OF ACETOLYSIS OLIGOSACCHARIDES CONDITION IIIb FROM P. past& IF0 0948 STRAIN MANNAN Relative retention time 1.00 1.66 1.88 2.21
W!
M8
1.00 0.83 +42.1
OBTAINED
M4
1.00 -
-
1.87 Trace
-
+14.4
UNDER
MS
1.00
1.00 -
2.69
4.01 -
-17.2
-0.5
502
KOBAYASHI,
SHIBATA,
FIG. 6. Time course of acetolysis of mannotetraose (0) and mannopentaose (0) with a 10/10/l (v/v) mixture of (CH&O)cO, CHaCOOH, and HaSO, at 40°C. These oligosaccharides were obtained from the P. pastoris IF0 0948 strain mannan by acetolysis with a 109/199/l (v/v) mixture of (CH&O)aO, CH&OOH, and H&SO, at 40°C for 36 h. Degradation ratios of oligosaccharides were calculated with the formula described under Materials and Methods.
notriose, and mannopentaose upon partial acetolysis under the same conditions as those of Condition Ib, and that the chemical structures of mannobiose and mannopentaose were ManpcYl - 2Man and Manpal - 2Manppl 2Manp/31 2Manpal - ZMan, respectively. However, the results obtained in the present study indicated that the molar ratio of the oligosaccharides obtained by acetolysis under Condition Ia was closely identical to that of the oligosaccharides obtained under Condition Ib. Therefore, it was reasonable to consider that Condition Ib was too drastic to leave /3-1,2-linked manno-oligosaccharides as well as Condition Ia. This finding also suggested that the decrease in H&SO4 concentration in acetolysis medium might be effective for controlling the acetolytic rate. We therefore attempted an investigation to find out the proper acetolysis condition for the isolation of manno-oligosaccharides containing p-1,2 linkage without serious degradation. One of the significant results obtained in the present study was the increase in the amount of mannotetraose under Condition III which was scarcely produced under Conditions Ia and -b from the same mannan. The results of PMR, specific rotation,
AND
SUZUKI
and methylation studies provided evidence that the structures of mannotetraose and mannopentaose were Manpbl 2Manpfll - 2Manpcwl - 2Man and Manpcrl - 2ManpPl 2ManpPl 2Manpal - ZMan, respectively. It is also of interest that the difference in the susceptibility of these oligosaccharides to acetolysis is considerably large, providing evidence that the nonreducing terminal sugar residue connected to the p-1,2 linkage is more sensitive to acetolysis than that located in the intermediary position connected to the same linkage. This finding seems to agree with the findings reported by Zhang and Ballou (25) on the prompt degradation of fi-1,2-linked manno-oligosaccharides with a 10/10/l mixture of (CH&O)PO, CH&OOH, and H,SO,; i.e., the relative acetolytic rate of Manpoll - BMan, Manp@l - BMan, and Manpal - 6Man was approximately 30/2/l. Additionally, it was observed in the present study that, even under a mild acetolysis condition such as Condition III, the degradation of Manppl - 2Manppl - 2Manpcrl - 2Man took place at a measurable rate. Therefore, depiction of the chemical structure of the mannan of P. p&or& IF0 0948 strain on the basis of findings from acetolysis studies should be made by taking account of the high susceptibility of Manppl - 2ManpPl + 2Manpal - 2Man to acetolysis, even milder conditions were adopted to the acetolysis.
3 4 FIG. 7. Representative structure of the P. pasti IF0 0948 strain mannan. M denotes a D-mannopyranosyl residue. The numbers given outside the brackets indicate the approximate number of side chains, the sequences of which are not specified.
ACETOLYSIS
OF
b-1,2
LINKAGE-CONTAINING
Summarizing all findings obtained by Gorin et al. (8) and by us, a representative chemical structure of the mannan of P. pastoris IF0 0948 strain can be depicted as shown in Fig. 7. It is obvious that the longest side chain in the P. pastoris IF0 0948 strain mannan was Manwl - 2Manpal 2Manpfil - 2Manpcyl - BMan, although the predominant oligosaccharide detected in the acetolysate was Manp/31 2ManpPl - 2Manpal - BMan, which was found to be about 2 times larger than the former in the molar ratio. Because the content of alkali-eliminable oligosaccharide residues in P. pastoris IF0 0948 strain mannan was found to be low, less than 1.0% (data not shown), it is reasonable to state that these oligosaccharide residues can be disregarded as a depiction of the chemical structure of the mannan moiety. Conclusively, the use of acetolysis medium containing H2S04 at a relatively low concentration, up to 0.5% (v/v), in which occurs the preferential cleavage of the a1,6 linkage in a yeast mannan giving rise to manno-oligosaccharides containing acetolysis-labile /l-1,2 linkage seems to be advantageous for the structural investigation of mannose-containing polysaccharides and high-mannose-type carbohydrate moieties of many glycoproteins. REFERENCES 1. SUMMERS, D. F., GROLLMAN, A. P., AND HASENCLEVER, H. F. (1964) J. Immu?wL 92,491-499. 2. HASENCLEVER, H. F., AND MITCHELL, W. 0. (1964)
J. Immwwl
93,763-771.
3. LEE, Y., AND BALLOU, C. E. (1965) Biochemistry 4, 257-264. 4. SUZUKI, S., SUNAYAMA, H., AND SAITO, S. (1968) Japan J. Microbial 12,19-X 5. KOCOUREK, J., AND BALLOU, C. E. (1969) J. BacterioL 100,11’75-1181.
503
MANNAN
6. GORIN, P. A. J., AND PERLIN, A. S. (1956) Can&. J. 0x-m 34,1796-1803. 7. SUZUKI, S., AND SUNAYAMA, H. (1968) Japan. J. Microbial 12,413-422. 8. GORIN, P. A. J., SPENCER, J. F. T., AND BHATTACHARJEE, S. S. (1969) Canad. J. Cfiem 47,14991505. 9. HAMADA, T., NAKAJIMA, T., IZAKI, K., AND MATSUDA, K. (1981) Eur. J. B&hem 119,365-371. 10. FUNAYAMA, M., NISHIKAWA, A., SHINODA, T., SuZUKI, M., AND FUKAZAWA, Y. (1984) Micro&o1 Immunol 28,1359-13’71. 11. GORIN, P. A. J., AND SPENCER, J. F. T. (1970) Adw. Appl. Micr&iol 13,25-89. 12. DUBOIS, M., GIL.LES, K. A., HAMILTON, J. K., REBERS, P. A., AND SMITH, F. (1956) Anal Chem 28,350-356. 13. AMES, B. N., AND DUBIN, D. T. (1960) J. Biol. Chem 235,769-775. 14. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. Biol. Chem 193, 265-275. 15. WEILL, C. E., AND HANKE, P. (1962) Anal. Chem 34,1736-1737. 16. OKUBO, Y., SHIBATA, N., ICHIKAWA, T., CHAKI, S., AND SUZUKI, S. (1981) Arch B&hem. Biophys. 212,204-215. 17. ISHITANI, T., AND HIRATA, T., AND KATO, K. (1978)
Agri. Biol. Chem. 42,897-898. 18. OKUBO,
Y., ICHIKAWA,
T., AND SUZUKI,
S. (1978)
J.
Bacterial. 136, 63-68. 19. OKUBO, Y., AND SUZUKI, S. (1978) 62,135-141. 20. HAKOMORI, S. (1946) J. B&hem. 208.
Carbohydr. Res. (Tokyo) 55,205-
21. LINDBERG, B. (1972) in Methods in Enzymology (Ginsburg, V., ed.), Vol. 28, Part B, pp. 178-195, Academic Press, New York/London. 22. STEWART, T. S., AND BALLOU, C. E. (1968) Biochemistry 7,1855-1863. 23. JONES, G. H., AND BALLOU, C. E. (1969) J. Biol
Chem. 244,1052-1059. 24. COHEN,
R. E., AND BALLOU,
C. E. (1980) Biochem-
istry 19,4345-4358. 25. ZHANG, W., AND Bwou, 256,10073-10079.
C. E. (1981)
J.
Bid Chem