THE JOURNAL OB BIOLOGICAL CHEMISTRY Vol. 244, No. 4, Issue of February 25, PP. 1043-1051,1969

Printed in U.S.A.

Studies on the Structure of Yeast Mannan

I. PURIFICATION AND SOME PROPERTIES OF AN a-MANNOSIDASE FROM AN ARTHROBACTER SPECIES*

(Received for publication, July 11, 1968)

GEORGE H. JONES AND CLINTON E. BALLOU

From the Department of Biochemistry, University of California, Berkeley, California 94720

SUMMARY

A soil microorganism, designated Arfhrobacfer GJM-1, has been isolated which is capable of growing on Sac- charomyces cerevisiae mannan as the carbon source. When grown on mannan, the microorganism secretes an a-man- nosidase into the cultural medium. The enzyme is not found when mannose, glucose, or glycerol is substituted for mannan, but is present when oligosaccharides obtained by acetolysis of mannan are used as carbon sources. The cr-mannosidase has been isolated and partially purified. It was free of contaminating glycosidase, phosphatase, and protease activities.

The enzyme is active on mannans from several different yeasts and on the oligosaccharides produced by acetolysis of yeast mannans, but has a low activity on cr-mannosides with aromatic or aliphatic aglycons. This is in contrast to results obtained with other cr-mannosidases and may reflect the different metabolic roles of the respective enzymes.

The K, values for the di-, tri.-, tetra-, penta-, and hexasac- charides from S. cerevisiae mannan varied between 0.5 and 2 fll~. Because of the structural heterogeneity of these oligosaccharides, it is not certain what features contribute to the differences observed. The Km value for intact mannan was 1 X 10s6 M, but this value probably reflects the higher concentration of side chains in the mannan molecule.

Mannan is an important structural component of the yeast cell wall (1,2). The structure of the mannan from Xuccharomyces cerevisiue, or baker’s yeast, has been studied by methylation (3) and by the degradative technique of acetolysis (4, 5). With acetolysis, a series of oligosaccharides containing o-(1 -+ 2)-, a-(1 -+ 3)-, and cr-(1 + 6)-linkages has been isolated from S. cerevisiae mannan. The acetolysis and methylation data indicate

* This work was supported by Grants AM8845, AM10109, and TI-GM31 from the United States Public Health Service and Grant GB-5566 from the National Science Foundation. It was taken from the doctoral thesis of George H. Jones.

that the mannan is a highly branched polysaccharide with cr-(1 --f 2)- and a-(1 -+ 3)linked side chains attached to an cr-(1 + 6)linked backbone (5). No evidence for a simple repeat- ing unit in mannan has been advanced.

Mannans have been isolated from the cell walls of many yeasts as well as from cultural filtrates and from the cell sap of some species (2, 6, 7). These mannans usually contain phosphorus and nitrogen (2). The phosphorus in mannans is diesterified and phosphorus-containing oligosaccharides have been isolated from acetolysates of Kloeckera brewis mannan (8). The nitrogen is present as hexosamine and protein (1,2). Although there are reports of the isolation of mannan-protein complexes from yeast cell walls (9, lo), not much is known about the way in which the carbohydrate and protein are associated. The enzyme, yeast invertase, is perhaps the best known example of a mannan- protein (7).

Most studies on the structure of mannans have been performed with chemical techniques. It was felt that a significant con- tribution to the knowledge of mannan structure could be made by combining chemical studies with enzymatic techniques. Several known cr-mannosidases are relatively inactive on yeast mannan (11, 12). Consequently, a search was instituted for an enzyme capable of digesting this polysaccharide. Such an enzyme has been obtained from a soil microorganism grown on a mannan-salts medium. This report describes the isolation, partial purification, and some properties of this enzyme. A preliminary account of this work has appeared (13).

EXPERIMENTAL PROCEDURE

Substrates-Mannan was prepared by citrate buffer extraction (14) of pressed baker’s yeast (Red Star Company, Oakland, California) or by extraction of fresh cells of K. brevis. The mannans of Candida albicans, Candida sfellatoidea, and Candida tropicalis were donated by Dr. H. F. Hasenclever, National Institutes of Health, Bethesda, Maryland. p-Nitrophenyl a-n-mannoside was a gift of Dr. Y. C. Lee, McCollum-Pratt Institute, The Johns Hopkins University, Baltimore, Maryland. Oligosaccharides were prepared from different mannans by acetolysis and purification of the products by gel filtration (4,5). See Fig. 1 for the structures of these oligosaccharides. The

1043

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

1044 Studies on Xtructure of Yeast Mannan. I Vol. 244, No. 4

M 2 M3 M; -M-4

M M 1

M 2

I 2

t

2 M 2

M’ M’ M’ I

t

M’ 2 2

M’

1 3

M’ I

M’ I 3

M’

Ms ISOMERS

M-M

M6 ISOMERS

t 2

M’

I --qM

7 I

2 2

M’ M’

I 2

M’

I

3

M’

I >M

i I

2 2

M’ M’

I 2

M’

I 3

M’

I LM

il I

2 2

M’

T I 2 2

M’ M’

FIG. 1. Structures of the oligosaccharides isolated from ace- tolysates of S. cerevisiae mannan. M denotes an a-n-mannopyr- anosyl residue. After Stewart, Mendershausen, and Ballou (5).

preparation of myoinositol 2-O-a-n-mannoside and of glyceryl- phosphorylmyoinositol dimannoside has been described (15, 16).

General Methods-Carbohydrate was determined by the method of Dubois et al. (17), protein by the method of Lowry et al. (18), and reducing sugar according to Paleg (19).

Descending paper chromatography was done on Whatman No. 1 paper. The following solvents were used (in volume ratios) : A, ethyl acetate-pyridine-water (5:3 :2); B, ethyl ace- tate-pyridine-water (10:4:3); C, ethyl acetate-acetic acid-formic acid-water (18 :3 : 1:4) ; D, I-butyl alcohol-pyridine-water (10:3:3). Sugar spots were detected with the silver nitrate- sodium hydroxide dip reagent (20).

Disc electrophoresis was performed according to Davis (21) with some modifications. Lower gels were prepared as described with ammonium persulfate to catalyze polymerization. The gels were then subjected to electrophoresis for 1 hour at a current of 4 ma per tube. The buffer compartment,s contained the Tris- chloride buffer used to prepare the lower gel. This treatment served to eliminate artifacts produced by the persulfate catalyst. Concentrated sucrose solutions were used in place of the upper gel. Equal volumes of enzyme solution and 2 M sucrose in 0.1 M

potassium phosphate buffer, pH 6.8, were mixed and layered on the lower gel. The Tris-glycine buffer was carefully layered over this solution and electrophoresis carried out. Gels were stained with 1 y0 Amido black in 7 y0 acetic acid.

Oligosaccharides were acetylated in a 1:l mixture of anhy- drous pyridine and acetic anhydride, 10 to 50 mg of saccharide in 5 ml of the mixture, by heating on a steam bath for 12 hours. The solvent was evaporated and the residue extracted with chloroform, which was washed three times with water and then evaporated to dryness. The acetylated oligosaccharide was dissolved in 2 ml of acetolysis medium, a mixture of acetic acid, acetic anhydride, and concentrated sulfuric acid in the ratios lO:lO:l, v/v. The solution was warmed in an oil bath at 40”. To obtain complete cleavage of a-(1 -+ 6)linkages, acetolysis was carried out for 13 hours (5). The reaction was stopped by the addition of 2 volumes of anhydrous pyridine to the solution. The solvent was evaporated and the oily residue was taken up in chloroform. The chloroform extract was washed three times with water and the water wash was extracted three times with chloroform. The combined chloroform extract was dried over anhydrous sodium sulfate and evaporated to dryness. The residue was dissolved in a few milliliters of anhydrous methanol and barium methoxide in methanol was added dropwise until the solution became alkaline. After 15 min, solid CO2 was added to neutralize the base and the methanol was evaporated. The deacetylated oligosaccharides were dissolved in 2 ml of water and the barium carbonate which formed was removed by centrifuga- tion. Deacetylated acetolysis products were separated by gel filtration.

Enzyme Assay-The assay mixture contained 250 pg of man- nan, 0.50 ml of 0.1 M potassium phosphate buffer (pH 6.8), 0.1 pmole of CaCla, 50 pg of bovine serum albumin, 1 to 3 pg of enzyme protein, and water to 1.0 ml. The mixture was incu- bated for 1 hour at 37”, when the reaction was stopped by the addition of 1.0 ml of the alkaline copper reagent described by Paleg (19). The tubes were heated in a boiling water bath for 25 min. After the tubes cooled, 1.0 ml of the arsenomolybdate reagent was added, the tubes were stirred vigorously, and the contents were diluted to 10 ml. After 15 min, the absorbance was read at 560 rnp in a Zeiss PM& II spectrophotometer. Boiled enzyme or zero time controls were subtracted from the sample values. The amount of reducing sugar released was determined from curves with mannose as the standard. One unit of activity was defined as the release of 1 pmole of mannose eq per hour under the assay conditions.

Enzyme samples were assayed in triplicate with zero time or boiled enzyme blanks for each determination. The assay was linear up to 100 pg of reducing sugar. Enzyme samples were diluted so that 25 to 65 pg of reducing sugar were released under the assay conditions. Phosphate buffer containing 1.0 mg per ml of bovine serum albumin was used as diluant.

RESULTS

Isolation and Identi$cation of Mannosidase-producing Microorganism

A soil microorganism which was capable of growing on yeast mannan was isolated by enrichment culture techniques (13). The organism was grown on a medium of the following composi- tion: 1 g of yeast mannan, 500 mg of (NH&S04, 200 mg of MgS04, 10 mg of FeS04.7Hz0,50 mg of CaCla.ZHzO, 500 mg of

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1969 G. H. Jones and C. E. Ballou

yeast extract, 7.54 g of K~HPOI, 2.32 g of KHzPO*, and water to 1 liter. The cells were grown at 30” with shaking, and the growth curve is shown in Fig. 2. It was necessary to incubate the cells for 36 hours to obtain maximal production of enzyme and digestion of mannan. Analysis of cultural filtrates and of harvested cells, which had been disrupted by sonic radiation, showed that 9570 of the total ol-mannosidase activity was in the cultural filtrate.

The mannosidase-producing microorganism was an obligate aerobe which did not form spores. The cells were long, slender rods in young cultures and were nonmotile. The long rods were replaced by shorter rods and coccoid shapes in older cultures. Stationary phase cells were 2 to 3 p in length and were about 1 p in width. Colonies of the microorganism were round and yellow and they showed pseudomycelial growth.

The microorganism was strongly gram-positive but was not acid-fast at any stage of growth between 0 and 96 hours of incubation. The cells grew well at 30” and 37” on media con- taining 0.5% glucose or mannose, supplemented with 0.05% yeast extract. The microorganism grew well on 0.5% yeast extract plus salts, and on the glycerol medium used for Myco- bacteria (22), but did not grow as well on the latter medium as on glucose, mannose, or mannan. The organism secreted a yellow pigment into the cultural medium which showed a bluish green fluorescence when irradiated with ultraviolet light.

Two liters of cells were grown on the glycerol medium used for Mycobacteria (22) and were extracted with 70% ethanol in a search for the 6-0-methylglucose-containing lipopolysaccharide found to be characteristic of several species of Mycobacteria (23). The extract was worked up as described by Saier and Ballou (24). Carbohydrate-containing fractions from the Sephadex G-50 column were pooled, concentrated, and hydrolyzed in 2 N

H$Ob at 100”. The neutralized hydrolysates were chromato- graphed on paper in Solvents B, C, and D. No spot corre- sponding to 6-0-methylglucose was observed. The hydrolysates contained substantial amounts of inositol and glycerol, but only a small amount of mannose, indicating that the mannophospho- inositides which are found in Mycobacteria, Corynebacteria, and Nocardia (23, 25, 26) were present in very small amounts, if at all. On the basis of these findings, the organism was identified as a member of the genus Arthrobacter, and was designated Arthrobacter GJM-1.

Isolation and Pur@cation of cr-Mannosidase

Cultural Filtrate-Five cultures of 1 liter each were grown on the mannan-salts medium described above. All subsequent steps were carried out at 4”. The cells were harvested after 36 hours by centrifugation in a Sorvall automatic refrigerated centrifuge at 10,000 rpm for 30 min. The cultural filtrate was dialyzed for 24 hours against 0.01 M potassium phosphate, pH 6.8, containing 10W4 M Ca* (Buffer A). Aliquots, 500 ml, of the cultural filtrate were placed in suitable lengths of cellulose dialysis tubing (1: inch diameter, Union Carbide). Five such sacs were placed in a 15-liter bell jar containing 10 liters of Buffer A. The dialysate was replaced with fresh buffer after about 12 hours. Finally, the solution in the dialysis sacs was lyophilized in 2.5- liter batches.

Ammonium Sulfate Precipitation and Back Extraction-The dried enzyme from the previous step was dissolved in 0.1 M potassium phosphate buffer, pH 6.8, containing 10V4 M Ca* (Buffer B) to give a final protein concentration of about 2.5 mg

1.00 I I .80

.60

I .40

t

=I z

.06

.04

‘h., --..

‘h

2.00

1 T

l.OO- $

.80- s

B .60- _ -2

E .40- z

Y -8

f,

.20- 5 9

.Ol II I I I ! I 0 6 12 18 24 30 36

TIME (HOURS)

1045

1000 800 ‘i-

80 5

r”

10

FIG. 2. Growth curve of the a-mannosidase-producing micro- organism. One liter of the cultural medium was inoculated with 10 ml of an l&hour liquid culture and grown in a conical flask at 30” on a New Brunswick gyrorotary shaker at 150 rpm. Aliquots were removed aseptically at intervals and analyzed for cell dry weight and for mannan and enzyme concentrations. Cell dry weight was determined on samples which had been dried at, 120” for 15 hours, and is represented in the figure by the solid line. The dashed line represents the mannan concentration (milligrams per ml) at various stages of growth and the dotted line represents the enzymatic activity (micrograms of mannose equivalent re- leased per hour by 1 ml of cultural medium).

per ml. About 200 ml of buffer were used. The enzyme was precipitated from this solution by the addition of solid am- monium sulfate (610 mg per ml). The protein suspension was allowed to stand at 4” overnight and then was centrifuged at 12,000 rpm for 30 min. The resulting precipitate was extracted with 50 ml of a 60% ammonium sulfate solution. After standing for several hours at 4”, the precipitate was collected by cen- trifugation at 15,000 rpm for 30 min. Two centrifugations were necessary for complete precipitation of the enzyme.

Xephadex G-100 Gel Filtration-The ammonium sulfate pre- cipitate was dissolved in 10 ml of Buffer B to give a final protein concentration of about 20 mg per ml. This solution was applied in 2.5-ml batches to a column, 2 x 100 cm, of Sephadex G-100, previously equilibrated with Buffer B. The elution pattern from this column is shown in Fig. 3. The mannosidase-contain- ing fractions were pooled and concentrated to 25 ml in an Amicon model 400 ultrafilter at an argon pressure of 70 psi with the UM-1 membrane.

DEAE-cellulose Column Chrmatography-The concentrated Sephadex G-100 fraction was dialyzed overnight against 0.05 M Tris-chloride buffer, pH 9.0, containing 10V4 M Ca* (Buffer C) and the dialyzed protein was applied to a column, 1.8 x 12 cm, of DEAE-cellulose equilibrated with the same buffer. The

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

1046 Xtudies on Structure of Yeast Mannan. I Vol. 244, No. 4

FIG. 3. Chromatography of crude a-mannosidase (60% am- monium sulfate fraction) on Sephadex G-100. The column was loaded with 2.5 ml of enzyme solution (equivalent to 50 mg of pro- tein) and eluted with 0.1 M potassium phosphate buffer, pH 6.8, containing 10-4 M Ca++. The solid line represents the absorbance at 280 rnp (protein) and the dashed line the absorbance at 560 rnp (enzyme activity). The most active fractions were pooled and purified further.

0 25 50 75

FRACTION NUMBER

FIG. 4. DEAE-cellulose chromatography of a-mannosidase. The concentrated Sephadex G-100 fraction was dialyzed against 0.05 M Tris buffer, pH 9.0, containing lo+ M Ca++. The dialyzed solution was applied to a column, 1.8 X 12 cm, of DEAE-cellulose equilibrated with the same buffer. Elution was accomplished with a linear gradient of 0 to 0.6 M KCl. The solid line represents the absorbance at 280 rnp (protein), and the dashed line the ab- sorbance at 560 m,u (enzyme activity). The shape of the KC1 gra- dient is also depicted in the figure. Active fractions were pooled and concentrated.

enzyme was eluted with a linear gradient of 0 to 0.6 M KCI. The mixing chamber of the gradient device contained 250 ml of Buffer C, and the reservoir had 250 ml of 0.6 M KC1 in Buffer C. The elution pattern is shown in Fig. 4. Active fractions were pooled, dialyzed against Buffer B, and concentrated in a Amicon model 50 ultrafilter under a nitrogen pressure of 40 psi. The final protein concentration was about 2 mg per ml and the yield

A summary of the purification procedure is given in Table I. The enzyme was purified approximately lo-fold with about 40 % recovery. Although the DEAE-cellulose step resulted in a 50 % loss of activity, it removed an acidic, yellow protein contaminant from the enzyme preparation. This was the major protein con- taminant at all stages of the purification prior to the DEAE- cellulose column and it could not be shown to possess any en- zymatic activity. The degree of purification of the enzyme was not increased by substitution of Sephadex G-150 or G-200 for G-100 or by chromatography of the enzyme on hydroxylapatite, although both procedures could be carried out with good recovery of enzymatic activity.

Properties of Enzyme

The purity of the enzyme was assessed at each stage of the purification by disc electrophoresis on 8 y0 anionic polyacrylamide gels. The concentrated enzyme from the DEAE-cellulose column showed two protein bands which stained with about equal intensities. The two bands had very similar mobilities and were only 1 mm apart after 1.5 hours of electrophoresis. Attempts to elute the enzyme from the gel were only slightly successful. In these experiments a gel, to which 100 pg of protein had been applied, was cut in half after electrophoresis. One half was stained for protein with Amido black. The other half was cut into O&cm strips and each strip was placed in a small test tube containing 1 ml of a 0.1% solution of bovine serum albumin in Buffer B. The gel strips were macerated with a spatula and the suspensions were allowed to stand at room temperature for 1 hour. The gel fragments were removed by centrifugation and the supernatant liquid was dialyzed for 12 to 24 hours at 4” against Buffer B and then assayed. Only a small fraction ( < 5 7,) of activity was recovered by this technique, but other methods (e.g. freezing and thawing) resulted in even lower recoveries. The activity recovered was located on the same strip as the two protein bands. No other strip contained activity or protein. Since it was not possible to determine which of the two bands represented the enzyme, it may be only 50% pure.

The purified enzyme was free of other glycosidase activities (see below) and was not contaminated with phosphatase or protease activity. It digested mannan to the same extent as

TABLE I PurQication of a-mannosidase

step Volume

ml

Cultural filtrate. . . 4890 Dialysis and ly-

ophilization. . 245 60% Ammonium

sulfate precipi- tate............. 9.5

Sephadex G-100.. 130 DEAE-cellulose.. 150 Concentration. . 15

-

-

-7-

Protein

ml?

875

610

200 65 30 29

Lctivity

X10” unitsa

2.88

2.78

2.56 2.00 0.97 1.06

-

! a

-

Specific activity

33

46

128 308 323 366

-

c _-

-

Purifi- :ation’

1.4 97

3.9 89 9.3 69 9.6 34

11.0 37

Yield

%

100

Q One unit is defined as the release of 1 pmole of mannose eq per hour under the assay conditions described in the text.

b This column gives the increase in specific activity over that was 29 mg from 5 liters of cultural filtrate. of the cultural filtrate at each stage of the purification.

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1969 G. H. Jones and C. E. Ballou 1047

did the crude preparation and the digestion products were the same. Thus, no enzyme of a particular specificity was lost during the purification procedure.

pH Optimum and Protein, Substrate, and Time Dependence- The enzyme was active between pH 5 and 9 with an optimum between 6.5 and 7.0. Maximal activity was obtained in potas- sium phosphate buffers. The mannosidase activity was de- pendent on added enzyme, substrate, and time of incubation (Figs. 5 to 7). The added mannan was about 40% hydrolyzed after 3 hours of incubation.

Stability-The enzyme was stable to repeated freezing and thawing and could be stored at -10” indefinitely with only a slight decrease in activity. It was stable at 4” for at least 2 days and could be kept at room temperature under 1 drop of toluene for 1 day with only a 15% decline in activity. No loss of activity occurred when the enzyme was heated at 50” for 5

0.8, I I I I TABLE II

I I I 25 50 75 100

/&t PROTEIN ADDED

FIG. 5. Dependence of the activity of partially purified a-man- nosidase on added protein. Assay mixtures contained the following ingredients: mannan; 250 rg; CaC12, 10 rmoles; 0.1 M potassium phosphate buffer (pH 6.8), 0.50 ml; bovine serum albumin, 50 pg; and water to 1.0 ml. Incubation was carried out at 37” for 1 hour. The reaction mixtures were analyzed as described in the text under “Enzyme Assay.”

I I I

,ug MANNAN ADDED

FIG. 6. Dependence of the activity of a-mannosidase on added substrate. Assay mixtures were prepared as described in Fig. 5 except that the amount of substrate added was varied.

0 60 120 180

TIME (Minutes)

FIG. 7. Dependence of the activity of a-mannosidase on the time of incubation. Assay mixtures were prepared as described in Fig. 5. Tubes were removed at various times after incubation began and were assayed for the release of mannose.

Effects of divalent cations MZ activity of partially purified or-mannosidase

Experiment and cation

1. Ca++ . . . 2. Mg++ . 3. Zn++. _. 4. cu++....... 5.5 Ca++ . .

- I

--

Cation concentration

10-x 10-z 10-a

M

55" 128 119 0 24 69 0 0 0 0 28 33

62 104 87

T

I -

-

IO-'

111 89 39 43 81

- I

-

10-s

105 98 53 61 81

= The figures in the table represent the percentage activity obtained at a given ion concentration relative to that of a control taken as 100%. In Experiments 1 to 4, the control used was the activity of the mannosidase in the absence of added divalent cation. Thus, in Experiment 1, the mannosidase activity was 28% greater in the presence of lo+ M Ca++ than in its absence. Incubation mixtures contained: mannan, 250 pg; 0.05 M Tris- chloride buffer (pH 7.0), 0.50 ml; bovine serum albumin, 50 pg; divalent cation (as the chloride), 100 ~1, and water to 1.0 ml.

b In Experiment 5 the effects of Ca++ on an enzyme sample which had been dialyzed successively against EDTA and distilled water were examined. The control represented the activity of the undialyzed enzyme in the presence of 1OW M Ca++. The activity of the dialyzed enzyme was 40% less than that of the undialyzed enzyme.

min, whereas the enzymatic activity was completely destroyed by heating at 70” for the same length of time.

E$ects of Added D&lent Cation-The effects of Ca++, Mg++, Cu++, and Zn* on the mannosidase are shown in Table II. All cations were added as the chloride. Added KC1 was without effect. The activity of the partially purified enzyme was stimu- lated by added Gaff ions. In contrast to crude preparations, which had maximal activity in the presence of 10e4 M Ca*, the purified preparation exhibited maximum activity at 10-S M.

Therefore, this concentration of Ca++ was used in assays of the partially purified enzyme. The enzyme was inhibited by Mgff, Cu++, and Zn++. Zinc ion has been shown to stimulate the mannosidases of rat epididymis and jack bean meal (27).

To determine whether the purified enzyme required added divalent cation, an aliquot was dialyzed against 1O-4 M EDTA in 0.05 M Tris-chloride buffer, pH 7.2. After 24 hours, the

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

1048 Studies on Structure of Yeast Mannan. I Vol. 244, No. 4

dialysis medium was changed to 0.05 M potassium phosphate buffer, pH 6.8. Dialysis was continued for an additional 24 hours. Assay of the dialyzed enzyme showed t#hat about 40% of the activity had been lost. The activit’y was completely restored when the enzyme was assayed in the presence of 1O-2 M Ca++.

iDIGESTS

FIG. 8. Photograph of a descending paper chromatogram of the products of the mannosidaee digestion of oligosaccharides (ethyl acetate-pyridine-water, 5:3:2). The standards are ~(1 + 6)-linked mannobiose and the di-, tri-, tetra-, and pentasaccharides isolated from yeast mannan by acetolysis (5). See Fig. 1 for exact structures. The oligosaccharides obtained by mannosidase di- gestion result from the cleavage of single mannose residues from the substrates. In the case of Mt and Mb these products have structures identical with the corresponding oligosaccharide from veast mannan acetolvsates. Thus. M4 vields Mn and M, vields &I2 and mannose. M: digestion yields a-mixture-of di-, tri:, and tet8rasaccharidescontainingcu-(1 + 6)- aswell as a-(1 + 2)linkages. M’, is 0- a-n-mannopyranosyl-(1 + 6)-n-mannopyranose.

M’LM + M=M 2

II

M 2

,I

M

M’LM 2

t 1

2

I I M M

M12M

t

2

IV/

t

2

M’

2

II 1

2

M M’

t

2

M’

I mannosidase digestion

M’LM M=M 2

t I

2

+ I +

M’

t 2

M’ tetrasaccharide fraction

mannosidase digestion

M

+ 2 t M’

2

I, trisaccharide fraction M

FIG. 9. Schematic representation of a method for characterizing the products of a-mannosidase digestion of Ms. Tri- and tetra- saccharide fractions obtained by mannosidase digestion were ex- amined by acetolysis. If both nonreducing ends of MS were at- tacked by the mannosidase, the tetrasaccharide fraction would yield Mz, M,, and mannose on acetolysis. If only one nonreducing end is attacked by the enzyme, the acetolysis products would contain Mz, or Ma plus mannose, but not all three. Similarly, if all of the nonreducing ends of the tetrasaccharide were attacked, acetolysis of the trisaccharide fraction should yield Mz, mannose, and some MS if the terminal ol-(1 + 6)-linkage of one isomer is broken.

SpeciJicity of cr-Mannosidase-The enzyme was active on the mannans of different strains of S. cerevisiae and on the mannans of K. brevis, C. albicans, C. stellatoidea, and C. tropicalis. It was also active on the oligosaccharides obtained by acetolysis of yeast mannans and on a series of (Y-(1 -+ 6)-linked oligoman- nosides (di- to hexasaccharide). The preparation of the latter oligosaccharides and the mode of action of the enzyme on intact mannans are described in an accompanying paper (28).

The a-mannosidase hydrolyzed p-nitrophenyl cr-n-mannoside at a very slow rate, and methyl cr-n-mannoside, glycogen, amy- lose, maltose, and cellobiose not at all. It was also inactive on a sample of crude Mycobacterium phlei mannolipid (22) and on myoinositol 2-C-cr-n-mannoside and glycerylphosphorylmyoino-

sitol dimannoside from this mannolipid fraction.

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1969 G. H. Jones and C. E. Ballou

/l&on of Enzyme on Oligosaccharide Substrates

Oligosaccharides from S. cerewisiae ilfannan-Assay mixtures were scaled up to 3 ml, with 200 pg per ml of a given oligosac- charide as substrate instead of mannan. Enzyme (6 pg) was added and the mixture was incubated for 1 hour, when the reaction was stopped by boiling. Two milliliters of the mixture were used to assay for the release of reducing sugar. The re- maining 1 ml was desalted by shaking it with mixed bed resin (carbonate form), after which it was concentrated and applied to Whatman No. 1 paper for chromatography with Solvent A. The results of this experiment are shown in Fig. 8. Reducing sugar was released from all of the oligosaccharides, although at different rates. Fig. 8 shows that the products of digestion were mannose, starting material, and the oligosaccharides which would be produced by cleavage of mannose residues from the starting material. For example, Mt and xl’2 yielded mannose only, M3 yielded Mz and mannose, and Md yielded mainly MS and mannose. M5 is a branched oligosaccharide composed of two structural isomers (see Fig. 1). Digestion of Mg yielded a mixture of products with the chromatographic mobilities of di-, tri-, and tetrasaccharides. These oligosaccharides contain cu-(1 -) 6)- as well as a-(1 -P 2)-linkages and are produced by sequential cleavage of single mannose units from both nonreduc- ing termini of the starting material. The exact structures of these products are described below. At times much shorter than 1 hour, MS digestion yielded only the tetrasaccharide fraction and mannose, confirming that the lower oligosaccharides were produced by stepwise cleavage of mannose residues from the starting material. When the reactions were allowed to proceed for 6 hours with additions of enzyme (6 pg each time) at a-hour intervals, each oligosaccharide was completely degraded to mannose. Thus, the enzyme can hydrolyze (1 --) 2)-, (1 -+ 3)-, and (1 -+ 6)-linkages in small substrates. Boiled enzyme con- trols showed only starting material on paper chromatography.

0.5

I

-I----.----I

M3 A 0.4 -

II i-J2

2 0.4 2

I I

[r M2 B

2 0.3 <

M 0.2

HJI I 0.1

I I

150 175 200

FRACTION NUMBER

FIG. 10. Separation on Sephadex G-25 of the products of ace- tolysis of the tetra- (A) and trisaccharide fractions (B) produced by mannosidase digestion of Ms. Column fractions were analyzed by the phenol-sulfuric acid method of Dubois et al. (17).

1/v 0,2K , , , 0,1[ My&. styllatoidea )

1 2 3 2 4 A

I I I I I 10 20 30 40 -k---- 20 30

0.11 , Mannpn (3. ;erevis;ae)

.05 .lO .15 .20

FIG. 11. Lineweaver-Burk plots of the kinetic data for S. cerevisiae mannan, Mz, M3, Mb, M, MC, and Md from C. stellatoidea mannan. Details of the incubations are given in the text. Sub- strate concentrations are expressed in millimoles per liter and velocities as micromoles of mannose equivalent released per min per mg of enzyme protein.

Characterization of Products of Mg Digestion-As mentioned above, mannosidate digestion of MS produced a mixture of lower oligosaccharides. These compounds were examined to determine whether the enzyme was capable of attacking both nonreducing ends of the M5 molecule. This was done by acetolysis of the tri- and tetrasaccharide fractions produced by digesting MS with cr-mannosidase. The products expected from the digestion are shown in Fig. 9. I f both nonreducing ends of MS were attacked, three different tetrasaccharides would be produced, two con- taining mannose linked a-(1 ---f 6)- to MS and one containing two disaccharides joined by an a-( 1 -+ 6)-linkage. Acetolysis of the tetrasaccharide fraction would yield MS, MS, and mannose. If only one of the nonreducing ends of Mr, were attacked by the enzyme, the acetolysis products would contain MP, or h!l) and mannose, but not all three. Similarly, if all of the nonreducing ends of the tetrasaccharide intermediates were attacked, acetoly- sis of the resulting trisaccharide fraction should yield M2, man- nose, and some M3 if the terminal a-( 1 -+ 6)-linkage of one isomer is broken (Fig. 9).

To test the above hypothesis, 10 mg of MS were digested for 1 hour with a limiting amount of enzyme. The reaction was stopped by boiling the mixture and the digestion products were isolated by gel filtration. The products obtained were starting

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

1050 Xtudies on Structure of Yeast Mannan. I Vol. 244, No. 4

TABLE III

Kinetic parameters of oligosaccharides of yeast mannan

Substrate Km VI&IX

M2 MS M4 M5 pJI6 MP

Mannan

mmozes/1iter /.kmoles mannose/ minlmg enzyme protein

2.3 15 1.9 14 0.4 5.1 0.5 13.6 0.4 10.6 6.9 29.0 1.0 x 10-Z 5.4

Q Isolated from C. stellatoidea mannan.

material and mono- to tetrasaccharide fractions. The tri- and tetrasaccharide fractions were purified by rechromatography and were then acetolyzed. The deacylated acetolysis products gave the elution pattern shown in Fig. 10. Ma, Mz, and mannose were obtained from the tetrasaccharide fraction, indicating that both nonreducing termini were attacked in Mg. Mt and mannose were produced from the trisaccharide but only a small amount of MS is observed. Thus, although the enzyme split both terminal (1 -+ 2)linkages in the tetrasaccharide, the terminal (1 -+ 6)- linkage was hydrolyzed at a much slower rate.

Kinetic Studies of Action of Lu-Mannosidase on Oligosaccharides of X. cerevisiae Mannan-The K, and V,,, values for the baker’s yeast mannan oligosaccharides were determined with the follow- ing assay mixtures: 0 to 500 pg of substrate, 0.5 ml of 0.1 M pot’assium phosphate buffer (pH 6.8), 10 pmoles of CaC12, 1.72 pg of enzyme protein, and water to 1.0 ml. After incubation for 15 min at 37”, reducing sugar was determined. Lineweaver- Burk plots of the data are given in Fig. 11 and the K, values are listed in Table III.

DISCUSSION

The enzyme described in this report is very active on yeast mannan. Many oc-mannosidases are known and, of these, the enzymes of rat epididymis, jack bean meal, and marine gastro- pods have been studied extensively (11, 12, 29). Significant differences exist between these enzymes and the bacterial enzyme described in this paper. One notable difference is that the bacterial enzyme was not very active on oL-mannosides with aromatic aglycons, whereas these compounds are good substrates for the other three enzymes. This finding could be related to the different metabolic roles of the various enzymes. If, for example, the rat, jack bean, and gastropod enzymes are involved in glycoprotein metabolism, they may encounter the aromatic side chains of tyrosine, tryptophan, and phenylalanine in regions near the glycosidic linkage to be hydrolyzed. Thus, their specificity could be such that these aromatic residues would not interfere. The Arthrobacter enzyme, on the other hand, was elicited presumably for the purpose of digesting yeast mannan. The glycosidic linkages in this substrate are adjacent almost exclusively to other polar mannose residues. Thus, the man- nosidase might be expected to be less active on mannosides con- taining nonpolar aglycons.

Another significant difference between the bacterial enzyme and the other three has to do with the effects of cations on their activity. Most reports of studies on the rat, gastropod, and

jack bean enzymes do not indicate any dependence on, or stimu- lation by, added divalent cations. It has been reported re- cently, however, that zinc ion stimulates the rat and jack bean enzymes, prevents their inactivation at 37”, restores activity which has been lost by incubation, and counteracts the inhibitory effecm of Cu++ and Cd++ (27). As can be seen from Table II, zinc ion was a potent inhibitor of the Arthrobacter enzyme at concentrations as low as 1O-5 M, and did not stimulate at any concentration tested. Calcium ion stimulated both the crude and purified enzyme and was able to restore activity which had been lost by dialysis against EDTA. This finding implicates the divalent cation in the mechanism of action of the enzyme. Calcium ion inhibits the mannosidase of the marine gastropod, Charonia lampas (11).

Studies with the oligosaccharides obtained by acetolysis of S. cerevisiae mannan have shown that the enzyme cleaves a-(1 ---f a)-, or-(1 -+ 3)-, anda-(1 ---f 6)-linkages in these substrates, releasing mannose residues from their nonreducing ends. It was surprising to find that the enzyme was not able to hydrolyze myoinositol 2-O-oc-D-mannoside since this compound is very similar in structure and presumably possesses the same conforma- tion as Mz (30). This finding further suggests that the enzyme is very sensitive to the structure of the aglycon.

Because of the structural heterogeneity of Ma, MC, and MG, it was not possible to define exactly the factors which contribute to the differences in K, and V DLaX obtained for the oligosac- charides. It does seem likely that the terminal a-(1 + 3)linkage is responsible for the low V,,, observed for Mq. It is also probable that the smaller K, values of MS and Me result from a combination of increased molecular size and the introduction of the branch points.

The K, value for intact S. cerevisiae mannan is 1 X 1O-5 M, based on a molecular weight of 22,000 (13). However, the molar concentration of side chains in the mannan molecule is about 50 times higher than the concentration of the mannan itself so that many more end groups are available for enzymatic attack. When the K, for mannan is multiplied by this factor, a value of 5 x lop4 M is obtained, which is close to those deter- mined for the oligosaccharide substrates.

In the accompanying paper (28), the mode of action of the enzyme on intact mannans from S. cerevisiae and other yeasts is described.

REFERENCES

1.

2. 3.

4.

5.

6.

7.

8.

9.

10.

11.

NORTHCOTE, D. H., AND HORNE, R. W., Biochem. J., 61, 232 (1952).

MILL, P., J. Gen. Microbial., 44, 329 (1966). CIFONELLI, J. A., AND SMITH, F., J. Amer. Chem. Sot., 77,

5682 (1955). LEE, Y. C., AND BALLOU, C. E., Biochemistry, 4, 257

(19651. STEWART, T. S., MENDERSHAUSEN, P. B., AND BALLOU, C. E.,

Biochemistry, 7, 1843 (1968). GORIN, P. A., AND PERLIN, A. S., Can. J. Chem., 34, 1796

(1956). NEUMANN, N. P., AND LAMPEN, J. O., Biochemistry, 6, 468

(19671. \ - - - - I -

STEWART, T. S., AND BALLOU, C. E., Biochemistry, 7, 1855 (1968).

KERN, b. D., AND NORTHCOTE, D. H., Biochem. J., 76, 12 (1960).

EDDY, A. A., Proc. Roy. Sot. (London), Ser. B.. 144, 425 (1958).

MURAMATSU, T., J. Biochem. (Tokyo), 62, 487 (1967).

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

Issue of February 25, 1969 G. H. Jones and C. E. Ballou 1051

12. LI, Y.-T., J. Biol. Chem., 242, 5474 (1967). 22. LEE, Y. C., AND BALLOU, C. E., Biochemistry, 4, 1395 13. JONES, G., AND BALLOU, C. E., J. Biol. Chem., 243, 2442 (1965).

(1968). 14. PEAT, S., WHELAN, W. J., AND EDWARDS, T. E., J. Chem. Sot.,

3862 (1968). 15. BALLOU, C. E., VILEAS, E., AND LEDEREIZ, E., J. Biol. Chem.,

238, 69 (1963). 16. LEE, Y. C., AND BALLOU, C. E., J. Biol. Chem., 239, 1316

(1964). 17. DUBOIS, M., GILLES, K. A., HAMILTON, J. K., REBERS, P. A.,

AND SMITH, F., Anal. Chem., 28, 350 (1956). 18. LOWRY, 0. Hi, R~SEBROUGH, fi. J., FARM, A. L., AND RANDALL,

P. J.. J. Biol. Chem.. 193. 265 (1951). 19. PALEG,‘L. G., Anal. Chem.,.Jl, 1912 (i959). 20. ANET, E. F. L. J., AND REYNOLDS, T. M., Nature, 174, 930

(1954). 21. D.~vIs, B. J., Ann. N. Y. Acad. Sci., 121, 404 (1964).

23. LEE, Y. C., AND BALLOU, C. E., J. Biol. Chem., 241, 1899 (1966).

24. SAIER, M. H., AND BALLOU, C. E., J. Biol. Chem., 243, 992 (1968).

25. ASSELINEAU, J., Biochim. Biophys. Acta, 64, 359 (1961). 26. LANELLE, M. A., ASSELINEAU, J., AND CASTELNUONO, G., Ann.

Inst. Pasteur, 108, 69 (1965). 27. SNAITH, S. M., AND LEVVY, G. A., Biochem. J., 106, 53P

(1968). 28. JONES, G. H., AND BALLOU, C. E., J. Biol. Chem., 244, 1052

(1968). 29. CONCHIE. J., FINDLAY, J., AND LEVVY, G. A.. Biochem. J.,

71, 318’(1959). ’ 30. BALLOU, C. E., AND LEE, Y. C., Biochemistry, 3, 682

(1963).

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from

George H. Jones and Clinton E. BallouSPECIES

-MANNOSIDASE FROM AN ARTHROBACTERαPROPERTIES OF AN Studies on the Structure of Yeast Mannan: I. PURIFICATION AND SOME

1969, 244:1043-1051.J. Biol. Chem. 

  http://www.jbc.org/content/244/4/1043Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/244/4/1043.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 25, 2019

http://ww

w.jbc.org/

Dow

nloaded from