THE JOIIRNAL OF BIOLOGICAL CHEMISTRY Vol. 237, No. 2, February 1962

Printed in U.S.A.

Alginic Acid Metabolism in Bacteria

I. ENZYMATIC FORMATION OF UNSATURATED OLIGOSACCHARIDES AKD 4-DEOXY-L-ERYTHRO-5-HEXOSEULOSE URONIC ACID

JACK PREISS AND GILBERT ASHWELL

From the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, United States Public Health Service, Bethesda 14, Maryland

(Received for publication, August 17, 1961)

Alginic acid, a high molecular weight linear polymer obtained from seaweed, is widely used as an added product to food due to its property of forming highly viscous solutions. Originally de- scribed as a polymer of n-mannuronic acid, it has now been dem- onstrated, by the chemical studies of Fischer and DGrfel (l), to be a ,&l-4-linked polymer of n-mannuronic and L-guluronic acid. The ratio of the two uranic acids vary widely in preparations of alginic acid derived from various species of seaweed. The al- ginic acid of Macrocystis pyr.ifera, the species examined in the present study, has been reported to have an L-guluronic acid content of 20 to 40% (2). At the present time, little is known of the sequence of the two uranic acids in the polymer.

In an attempt to delineate the enzymatic mechanism of alginic acid metabolism, as a prelude to studies on sequence determina- tion, an organism capable of growth on alginic acid as sole car- bon source was isolated. The results of metabolic studies with this organism have been reported briefly (3). According to the reaction sequence illustrated in Fig. 1, alginic acid is cleaved by an enzyme fraction referred to as alginase into a series of oligo- saccharides containing an unsaturated uranic acid on the non- reducing end of the oligosaccharide chain. These unsaturated oligosaccharides are further degraded with the ultimate produc- tion of the monosaccharide, 4-deosy-L-erythro-5-hexoseulose uranic acid.’

The present paper is concerned with the purification and prop- erties of alginase and with further documentation on the identity of the reaction products.

The following paper (4) describes the purification of a TPNH- linked dehydrogenase, from alginate-induced Pseudomonas, capable of converting the alginase end product, 4-deoxy-L-eryth- ro-8hexoseulose uranic acid, to 2-keto-3-deoxy-n-gluconate.

EXPERIMENTAL PROCEDURE

Various chemicals used in this study were generously made available to us as follows: n-Mannuronic acid by Dr. H. Isbell of the National Bureau of Standards, L-guluronolactone by Dr. F. G. Fischer, the c11- and P-metasaccharinic acids by Dr. H. Fletcher of this Institute, and alginic acid by the Kelco Com- pany, Los Angeles, California. Chondroitin sulfates A and B and hyaluronic acid were isolated and purified by Dr. Jean Hickman in this laboratory. Chondroitin sulfate C and poly- galacturonic acid were purchased from the California Corpora- tion for Biochemical Research and the Mann Research Labora-

1 This compound is also referred to as 4-deoxy-5-ketouronic acid.

tories, respectively, and used without further purification. All other reagents were from commercial sources.

Paper Chromatography Solvents-The solvents used in descend- ing paper chromatography on Whatman No. 1 or No. 3 filter paper were: n-butanol-acetic acid-water (50: 12:25), irrigation time, 48 hours (Solvent A) ; ethylacetate-acetic acid-formic acid- water (90 : 15 : 5 : 20), irrigation time, 6 hours (Solvent B) ; phenol- water (90 g to 10 ml), irrigation time, 18 hours (Solvent C); n-butanol-acetic acid-water (40 : 10: 50)) upper layer, irrigation time, 18 hours (Solvent D). The chromatograms were routinely developed by the silver nitrate method (5) or by spraying with periodate followed by thiobarbituric acid (6). In the case of Solvent C, phenol was removed by rinsing the paper twice in ether before drying and developing.

Cultivation and Harvest of Bacteria-An organism capable of utilizing alginic acid as sole carbon source was isolated from Potomac Bay mud by the enrichment culture technique.2 The organism, classified as a pseudomonad, was grown aerobically on a minimal salt medium composed of NH4N03, 1 %, K&fP04, 1.5%, NaHzPOh, 0.5 ‘%, MgS04.7H20,0.1%, and sodium alginate, 0.11%. After 5 days of growth at room temperature, the bac- teria were harvested in a Sharples centrifuge. The bacterial paste, bright orange in color, was stored frozen at - 15”. Approx- imately 1 g of cells (wet weight) was obtained per liter of culture fluid.

Enzymatic Assays

Three independent methods were used for measuring the ac- tivity of the enzyme, alginase, in degrading alginic acid. All three were proportional to protein concentration in the crude extract as shown in Fig. 2.

I. Or&no1 Assay-This method is based upon the stability of the undegraded alginic acid to brief heating in alkali. Both the monosaccharide end product and the intermediate reducing oligosaccharides give negligible orcinol reaction after alkali treat- ment.

A stock solution of sodium alginate, prepared by suspending 1.2 g of alginic acid in distilled water and adjusting the pH to 7.0 with 0.1 N NaOH, was brought to a final volume of 100 ml. This preparation contained an equivaient of 5i pmoles of uranic acid per ml when standardized against crystalline n-mannuronic acid. The reaction mixture, containing sodium alginate, 5.1 pmoles (mannuronic acid equivalents), KCl, 50 pmoles, Tris

2 The organism was iso!ated by Dr. J. D. Smiley.

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

310 Alginic Acid Metabolism Vol. 237, No. 2

ALGINIC ACID

-.. CHO

HOCH

HOCH = HOW I

HCH f=O

tOOH tOOH

4-DEOXY-S-KETO-URONIC ACID

FIG. 1. Enzymatic degradation of alginic acid

uv TEA

u 100 200 300 pGRAMS OF PROTEIN EXTRACT

FIG. 2. Linearity of the thiobarbituric acid (TBA), orcinol, and ultraviolet (UV) assay with respect to protein concentration.

buffer, pH 7.5, 30 pmoles, and enzyme in a total volume of 1.0 ml, was incubated for 8 minutes at 37”. The reaction was stopped by heating in a boiling water bath for 30 seconds and a 0.05 ml aliquot pipetted into 1.45 ml of 0.1 N NaOH. The latter tube was heated to 100” for 5 minutes, cooled, and assayed by the Dische modification of the orcinol reaction (7). A heated enzyme control was run with every assay. A unit is defined as that amount of enzyme required to cause a decrease of 20 units of optical density at 665 rnp in the total reaction mixture under the above conditions. The calorimetric assay was performed on a Beckman DU spectrophotometer with cuvettes with a l-cm light path.

II. Thiobarbituric Acid Assay-This method is based upon the formation of compounds giving rise to ,&formylpyruvic acid when subjected to periodate oxidation. Both 4-deoxy-5-keto- uranic acid and the unsaturated uranic acid moiety of the inter- mediate oligosaccharides are cleaved by periodate to form

@-formylpyruvate, whereas intact alginic acid does not so react. Thus, aliquots of the above described incubation mixture were assayed with thiobarbituric acid according to the modification of Weissbach and Hurwitz (8) whereby 0.01 pmole of P-formyl- pyruvate gave an optical density reading of 0.290 at 549 rnp. A unit is defined as that amount of enzyme required to yield 1 pmole of fi-formylpyruvate under the stated incubation condi- tions.

III. Ultraviolet Absorption Assay-This method is based upon the marked absorption peak at 230 mp of the unsaturated uranic acid portion of the intermediate oligosaccharides. Nei- ther intact alginic acid nor the final monosaccharide end product is measured in this assay. The same reaction mixture described in the previous two methods was used except that the concen- tration of sodium alginate was halved, i.e. 2.55 pmoles added (as mannuronic acid). The reaction was initiated by the addi- tion of an appropriate aliquot of the enzyme solution to a l-cm cuvette and the absorption determined at 230 rnp in a Beckman DU spectrophotometer. Readings, taken at 1 and 5 minutes after addition of the enzyme, were compared with a reaction blank in which the alginate was omitted. In this assay a unit is defined as that amount of enzyme required to cause an optical density increase of 1.0 at 230 rnp in 4 minutes at room tempera- ture. A similar assay has been reported by Nakada et al. (9) for measuring bacterial mucopolysaccharides.

Enzyme Purification

Crude Extract-The frozen bacterial paste was suspended in 4 volumes of 0.1 M potassium phosphate buffer, pH 7.5, containing 0.01 M glutathione, and disrupted by sonic vibration for 20 minutes in a lo-kc Raytheon oscillator. The broken cell mix- ture was centrifuged at 10,000 r.p.m. for 15 minutes and the supernatant solution was used as the starting material for puri- fication of the enzyme. All ensuing operations were carried out at O-3”.

Streptomycin Precipitation-Sixteen milliliters of a 5 y0 (weight per volume) streptomycin sulfate solution were added slowly, with continuous stirring, to 80 ml of the crude extract. After standing for 10 minutes, the suspension was centrifuged at 15,000 r.p.m. for 10 minutes and the supernatant retained (93 ml).

Ammonium Sulfate Fractionation-To 90 ml of the above frac- tion an equal volume of cold, saturated ammonium sulfate was added. After 10 minutes, the mixture was centrifuged as above, and the precipitate discarded. The supernatant was brought to 95% saturation by addition of solid ammonium sulfate and allowed to stand for 20 minutes before centrifugation. The re- sultant precipitate was dissolved in 20 ml of 0.03 M Tris buffer, pH 8.5, and dialyzed overnight against 500 ml of the same buffer (26 ml).

DEAE-Cellulose3-Twenty-five milliliters of the above solu- tion were added to a DEAE-cellulose column (2 X 12.5 cm) which had been equilibrated with 0.03 M Tris buffer, pH 8.5. Upon washing of the column with the same buffer solution, the enzyme appeared in the first two resin bed volumes. These were collected separately and designated DEAE-cellulose Frac- tions I and II (20 and 23 ml, respectively).

Table I summarizes the data on the partial purification of the alginate-degrading enzyme. The ratio of activity in the thio-

s Obtained from Serva Entwicklungs Labor, Heidelberg, Ger- many.

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

February 1962 J. Preiss and G. Ashwell 311

barbituric and ultraviolet assay remained constant with respect to each other during purification but changed appreciably in relation to the orcinol assay. Since each assay reflects different parameters of the same over-all reaction, the significance of this observation is not immediately clear. However, on the basis of preliminary experiments, it would seem reasonable to anticipate that more than one enzymatic activity is involved in the degradation of alginic acid.

Properties of Alginase

Xtability-After 3 months of storage at -15”: during which time the enzyme was repeatedly frozen and thawed, the activity decreased approximately 10 %; after 6 months, a decrease of 30 95 was noted.

pH Optimum-As measured by all three assays, alginase ac- tivity exhibited maximal velocity in the pH range of 7.0 to 8.0 (Fig. 3).

Salt Activation-The striking dependence of enzymatic ac- tivity upon salt concentration is illustrated in Fig. 4. Maximal activation was obtained at 0.05 M KCl; higher concentrations were inhibitory. A similar pattern of behavior was observed when assayed by the orcinol or thiobarbituric acid methods.

Specificity-Under conditions similar to those used in the al- ginic acid assay, hyaluronic acid, chondroitin sulfate A, B, and C, and polygalacturonate4 were inactive in the thiobarbituric acid or orcinol assays. Furthermore, no inhibitory effect of these compounds upon the enzymatic degradation of alginic acid was observed. Similarly, when the free uranic acids, n-mannuronate and L-guluronat,e, were incubated with the en- zyme, no detectable reaction occurred (thiobarbituric acid assay) and the uranic acids were recovered quantitatively. The in- ability of the enzyme to metabolize these compounds is in ac- cord with the concept that the free uranic acids are not involved as intermediates in the formation of 4-deoxy-5-ketouronic acid.

Isolation and Ident$cation of Reaction Products

The products found upon incubation of the partially purified enzyme with sodium alginate for 90 minutes at room tempera- ture were examined by paper chromatography in Solvent A. All of the components shown in Fig. 5 exhibited a strongly posi- tive reaction when sprayed with the thiobarbituric acid reagent (6) or treated with AgNOa (5). Corresponding areas of ultra- violet light absorption were observed in all cases with the excep- tion of the fastest moving compound. Prolongation of the in- cubation time to 24 hours failed to produce a qualitative change in the chromatogram, although a distinct increase in the staining intensity of the leading spot was apparent. On the basis of the relative rates of migration, as well as the differential response to thiobarbituric acid and ultraviolet absorption, it was presumed that the most rapidly moving component represented a mono- saccharide end product, whereas the slower moving components indicated the presence of a series of unsaturated oligosaccharides.

To accumulate an amount of the major reaction products sufficient for isolation and identification, a large-scale incubation was prepared. The reaction mixture, containing 2 mmoles of

4 Although the alginate-grown cell extracts were inactive with polygalacturonate as substrate, the organism grew readily with polygalacturonate as the sole carbon source. Extracts of the polygalacturonate-grown cells actively metabolized polygalac- turonate with the production of unsaturated oligosaccharides as

TABLE I

Puri$cation of alginase from Pseudomonas extracts

Total units

I I I /bi Crude extract 2,3404,8905,50013.0 Streptomycin supernatant 2,3204,9205,54011.3 Ammonium sulfate frac- 1,1001,7001,800 5.1

tion DEAE-cellulose Fractions :

I 504 693 800 0.3:

II 408 576 690 0.3!

1: iz -

Ratio of activities:

>I &Z

or&ml to thiobarbi-

- turic acid to ‘g 2

:: ultraviolet

units ld

2.81:2.1:2.3

2.71:2.1:2.4 8.61:1.5:1.0

78 1:1.4:1.6 29 1:1.4:1.7

* Orcinol assay.

lx 4F------T

0 6.0 6.5 7.0 7.5 8.0 8.5 9.0

PH

FIG. 3. Alginase assays as a function of pH. The assay proce- dares-are-describ.ed Under “W&ho&. “” Tlie bU&r useci,‘hhwever, in the pH 6 to 7 range was 0.02 M potassium phosphate instead of Tris-HCl.

I I I I I I

t

KCI

- *300 =L

-.OO .025 -05 .075 .I .I25 .I5

MOLARITY OF SALT

FIG. 4. Effect of salt concentration on alginase action. The ultraviolet assay procedure is described in “Methods” except

determined by the thiobarbituric assay. that the various salts were substituted for KC1 as indicated.

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

312 Alginic Acid Metabolism Vol. 237, No. 2

FIG. 5. Paper chromatography of the alginase reaction mixture. were chromatographed in Solvent A.

sodium alginate, 11 mmoles of Tris buffer (pH 7.5), and 4.5 mg of enzyme (DEAE-cellulose Fraction I) in a total volume of 75 ml, was incubated for 24 hours at 37”. During this time, 1,630 Mmoles of alginate (800/,) was converted to thiobarbiturate-re- acting products. The reaction mixture was diMed to 150 ml and placed on a Dowex 1-formate column (10% cross-linked, 4-cm diameter x 14 cm). The column was washed with 680 ml of water and elution begun with 0.15 N formic acid. Fractions (90 ml) were collected and assayed calorimetrically by the thio- barbituric acid procedure. The elution pattern is illustrated in Fig. 6. After collection of the major portion of the reaction product in Peak I (1.5 liters), the 0.15 N formic acid eluent was replaced by a linear gradient system containing 2 liters of water in the mixing chamber and 2 liters of 2.0 N ammonium formate in the reservoir. Two additional peaks were subsequently ob-

Twenty microlitera of the reaction mixture described in the text

tained. A total of 92% of the thiobarbiturate-reacting material was recovered in the three peaks.

The formic acid in Peak I was removed by ether extraction in a continuous flow liquid-liquid extractor as described previ- ously (10). Peaks II and III were treated similarly after pre- treatment with Dowes 50 (H+). All three fractions were con- centrated by lyophilization.

Identi&ation of Peaks I and II-When chromatographed in Solvents A and B, both peaks revealed a single thiobarbiturate- reacting component after periodate oxidation, which were in- distinguishable from each other in both solvent systems (Table II). The thiobarbiturate chromogen of both peaks yielded iden- tical spectra, closely paralleling that of known P-formylpyruvate, and indicated a recovery of 1,200 pmoles of product in Peak I and 200 pmoles in Peak II.

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

lcebruary 1962 J. Preiss and G. Ashwell 313

Subjection of both compounds to periodate oxidation failed to produce significant, amounts of formaldehyde as determined by chromotropic acid test (11). However, when the com- pounds were reduced with NaBH4 before periodate oxidation, a stoichiometric recovery of formaldehyde was found (Table II). Further supporting evidence for the presence of an intact alde- hyde group in the molecule was provided by the ulbraviolet spectrum of the semicarbazone derivatives (12) which exhibited a sharp maximum at 233 rnp (Fig. 7).

Both compounds gave a positive reaction when treated with o-phenylenediamine (13), thus inferring the presence of an a-keto acid. Similarly, both compounds were shown to be de- carboxylated readily in the presence of ceric sulfate (14) with the liberation of a stoichiometric yield of COZ (Table II).

Determination of periodate uptake, assayed spectrophoto- metrically (15), indicated a value of 1.93 moles of periodate con- sumed per mole of compound. The theoretical amount ex- pected from a 4-deoxy-5-ketohexuronic acid is 2 moles per mole of compound oxidized.

In an attempt to demonstrate the orientation of the hydroxyl groups on carbon atoms 2 and 3, the compounds from both Peak I and II were converted to the corresponding o(- and P- metasaccharinic acids by reduction with NaBH4 (Fig. 8). They were identified by their cochromatography in Solvents A and B with authentic crystalline standards. The reduction was shown to be quantitative by two assay methods: (a) determina- tion of formaldehyde released by pcriodate oxidation, and (b) measurement of lactone formation with FeCLhydroxylamine after a brief heating at 100” in 1 N HCl (16). The a- and @- metasaccharinic acids were then converted to 2-deoxyribose in 10 to 12% yields by the Ruff degradation (17) ; a control micro- degradation with authentic metasaccharinic acid resulted in a 15% yield.

The identification of the resulting products as 2-deoxyribose was based upon the following data: both products reacted to give the characteristic spectrum of authentic 2-deoxyribose in the cysteine-HzS04 (18) and diphenylamine (19) assay proce- dures. Repetition of the thiobarbituric acid assay no longer gave a maximum at 549 rnp but resulted in spectra identical to that shown by malonaldehyde with peak absorption at 534 mp

TABLE II Analytical data of Peaks I and II

Fraction

Peak I Peak II

&Formyl- Formaldehyde

pyruvate after HalOb:

oxidation Before After reduction reduction

0.800

0.600 w " 4' 0.500 z

5: 0.400 z

0.300

Migration Ceric sulfate Iecal;bo~yla-

soyt ,“g; B

pm&s/m1 cm/Z? cm/6 hr hr

20.8 26 8 15.5 26 8

L

210 250 300 350 MILLIMICRONS

FIG. 7. Absorption spectra of the deoxyketouronic acid semi- carbazone at a concentration of 4.4 X 10m5 M based on the thio- barbituric acid assay (7).

FHO CH20H CH20H :H20H

HO-F-H HO-F-H HO-F-H HO-C-H

HO-F-H NaBH4

HO-FH HO-F-H Ruff

HO-C-H

742 w CH2 + $H2

c

y=o HO-t-H

tH2

H-y-0” hH0

COOH LOOH COOH i=4 :: w 4.DEOXY-5KETO-URONIC a+~ META-SACCHARINIC 2.DEOXYRIBOSE

-3 ACID ACID

& FIG. 8. Chemical conversion of deoxyketouronic acid to 2-

22 deoxyribose.

. z (20). As can be seen in Fig. 9, 2-deoxyxylose can be distin-

s ' guished from 2-deoxyribose in this assay by their differing rates zz =L of oxidation with periodate. This rate difference, first demon-

0 strated by Waravdekar and Saslaw (20), has been attributed to 0 300 600 900 1200 1500 1800 2000 2400 2700 the more rapid oxidation of the adjacent cis-hydroxyls than the

MILLILITERS corresponding lruns configuration. The rates of oxidation of

FIG. 6. Chromatography of the alginate reaction products on the deoxypentoses derived from the compounds of Peaks I and Dowex I-formate. Procedure is described in the text. II are in close agreement with deoxyribose.

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

314 Alginic Acid Metabolism Vol. 237, No. 2

-REACTION

I I I I

0 IO 20 30 40

H,IO, OXIDATION TIME(MINUTES)

FIG. 9. Rate of malonaldehyde production from 2-deoxypen- toses. Malonaldehyde assayed by the thiobarbituric acid reac- tion. For each time point, 0.019 kmole of deoxypentose reaction product and 0.020 rmole of authentic deoxyribose and deoxyxylose based on the diphenylamine assay (18) were used.

I MILLIMICRONS

FIG. 10. Ultraviolet absorption spectra of the unsaturated oligosaccharide fraction (Peak III) at a concentration of 1.15 X It4 M.

Both compounds cochromatographed with 2-deoxyribose in Solvents A, B, C, and D, the last two of which clearly resolved 2-deoxyribose and 2-deoxyxylose. Optical rotation of the degra- dation product from Peak I was demonstrated to have an [oc]: of -40” which approximates the reported value of -50” for 2-deoxy-n-ribose (21) and establishes the former as the expected D isomer. The difference in the [al,” obtained from the renorted

5 Supporting evidence for the assignment of the tri- and tstra- saccharide structure to Fractions I and II, respectively (Table III), is based upon the sensitivity of the free reducing group of the terminal uranic acid to bromine oxidation and the apparent lack of reactivity of the unsaturated uronide moiety toward the orcinol reagent. Treatment of Fraction I with aqueous bromine at room temperature under conditions which completely destroyed the free uranic acid control resulted in a 50% loss of orcinol-reactingmate- rial, whereas the orcinol titer of Fraction II was decreased 33%. Since t,he nonreducing uronides are stable to mild oxidation, a decrease of 50 and 33% is in agreement with the theoretical values for unsaturated tri- and tetrasaccharides, respectively.

In addition, enzymatic degradation of Fractions I and II, by the crude extract, resulted in formation of close to the theoretical

. _ I L yield of 3 and 4 pmoles of deoxyketouronic acid, respectively.

TABLE III

Analytical data of oliyosaccharide fractions

Fraction Thiobar-

bituric acid

assay

Peak III

1 2 3

4

&m7les prmdes !J?PCOlCS

17.2 53.7 1.07 1.9 4.0 0.93

10.9 30.3 1.00 1.3 6.7 0.91 0.6 4.7 0.89

Orcinol Br2 uptake*

L*JL~,~ in 0.01 N HCI

8,500

-t 8,500 8,900

8,600 -

Orcinol thiobar-

bituric acid

2.1 19.7 2.8 12.0 5.1 7.4

7.8 4.2

Migration n Solvent

A

C?8/45 tr

* Bromine uptake was measured by the method of Suzuki (22) ; values are micromoles per pmole of thiobarbituric acid.

t Absorption from the paper was too high to permit an accurate estimation of the extinction coefficient of Fraction I.

value is attributed to the small amount of material available for measurement; insufficient material derived from the Peak II compound was available for a specific rotation determination.

On the basis of the above data, the compounds isolated in Peaks I and II appear to be identical and have been identified as 4-deoxy-L-erythro-5-hexoseulose uranic acid (Fig. 1). It is not apparent why this compound appears as two separate peaks upon Dowex-formate chromatography. However, in addition to the chemical evidence cited above, enzymatic similarity has also been demonstrated in that both compounds are converted to 2-keto-3-deoxygluconic acid by a purified TPNH-linked de- hydrogenase isolated from the same organism (4).

Characterization of Peak III-In contradistinction to the two fractions described above, Peak III exhibited a marked absorp- tion in the ultraviolet range with a sharp maximum at 235 rnp (Fig. 10).

When subjected to paper chromatography in Solvent A, Peak III was resolved into four separate components which were eluted from the paper. The data, compiled in Table III, indi- cate that each component possessed an extinction coefficient of 8,500 at 235 rnp and consumed 1 mole of bromine per mole of P-formylpyruvate released by pcriodatc oxidation. Comparison of the relative amounts of saturated to unsaturated uranic acid, based upon the orcinol to thiobarbiturate assay values, revealed a ratio which, in each case, approximated a characteristic inte- gral number. On the basis of this evidence, together with the behavior toward column and paper chromatography, it is sug- gested that these compounds represent the tri-, tetra-, hexa-, and nonasaccharides.6 Further analysis of these unsaturated oligosaccharides is in progress.

The observation that the unsaturated oligosaccharide gives

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

February 1962 J. Preiss and G. Ashwell 315

rise to P-formylpyruvate after periodate oxidation provides sug- gestive evidence for the location of the unsaturated bond at carbon atoms 4 and 5. From the studies of Waravdekar and Saslaw (20), it was shown that the release of malonaldehyde from galactal was more rapid than from 2-deoxyribose. No apparent reason was mentioned for this different rate. How- ever, as is illustrated in Fig. 11, galactal may be considered as the enol form of a 2-deoxy sugar just as the unsaturated uronide may be regarded as the enol form of a 2-keto-3-deoxycarboxylic acid. By analogy with galactal, P-formylpyruvate formation from the unsaturated oligosaccharide should be and is faster than from the corresponding 4-deoxy-5-ketouronic acid (Figs. 12 and 13).

DISCUSSION

The pattern of alginic acid utilization described in this study appears to be closely analogous to the bacterial degradation of mucopolysaccharides, as reported by Linker, Meyer, and Hoff- man (23) and Linker et al. (24). These workers isolated and identified a disaccharide composed of a A-4,5-uronide and Ar-acetylhexosamine which was subsequently shown to be enzy- matically hydrolyzed to yield 4-deoxy-n-three-5-hexoseulose uranic acid, the 2-epimer of the compound reported here.

A tentative mechanism proposed for the enzymatic degrada- tion of alginic acid is shown in Fig. 1. The polymer is pictured as being cleaved by an “endoalginase,” resulting in the forma- tion of a series of oligosaccharides containing an unsaturated uranic acid residue on the nonreducing end of the chain. The unsaturated oligosaccharides, in turn, are further fragmented with the eventual formation of 4-deoxy-n-erythro-hexoseulose uranic acid.

Provisional evidence in accord with the intermediate role of the unsaturated oligosaccharides is the finding of a series of slow moving, ultraviolet light-absorbing products after brief incuba- tion with minimal amounts of enzyme. Upon longer incubation, these products diminish in intensity and are accompanied by the appearance of the fast moving monomer. Under comparable conditions, n-mannuronic and L-guluronic acid are not metabo- lized as determined by the orcinol and thiobarbituric acid assays.

FHO HO-FH

H5106 CHO - Ht

$HO e HCH

HO-j;H :H5106

&-OH b=O 2

LO tOOH dOOH COOH COOH

UNSATURATED OLIGOSACCHARIDE

WI FHOH <HO $HO

He II

H5106 Hy - HO-tH -

-$H2 H5106 tH2 CHO CHO - HCOH

HO-+H 1 HtOH $H-0 MALONALDEHYDE tH20H

D-DEOXYRIBOSE GALACTAL

FIG. 11. Action of periodate on deoxyketouronic acid, n-de- oxyribose, gala&al, and unsaturated oligosaccharides.

k I I I I I “1 80 z

TETRA-SACCHARIDE

s -0 2 6 IO H5106 CONCENTRATION (XlO-3 M 1

FIG. i2. R&ease of B-formylpyruvate from tetrasaccharide and deoxyketouronic acid as a function of periodate concentration.

0 6 12 I8 24 30 36

H&O6 OXIDATION (MINUTES)

FIG. 13. Rate of release of fi-formylpyruvate from tetrasac- charide (O), trisaccharide (O), and deoxyketouronic acid (A).

The possibility of an alternate mechanism of degradation in- volving an “exoalginase” activity resulting in the progressive cleavage of monosaccharide units from the ends of the polymer has not been eliminated. However, the rapid formation of readily separable unsaturated oligosaccharides would tend to indicate that this is not a major constituent of the isolated al- ginase. An “exoalginase” attacking the nonreducing end of the polymer would be expected to form products reacting with thiobarbituric acid and absorbing ultraviolet light, whereas a similar attack on the reducing end of the chain would not be expected to give rise to unsaturated oligosaccharides.

As has been demonstrated, treatment of alginic acid with the purified enzyme leads to the formation of small amounts of un- saturated oligosaccharides which are not further degraded upon prolonged incubation. However, preliminary experiments have revealed that these compounds are rapidly metabolized by the crude extract, and it must be concluded that more than one enzyme is involved in the complete utilization of alginic acid. It is anticipated that further purification of the bacterial ex- tract, with the use of one or more of the resistant oligosaccha- rides as substrate, will provide sensitive and specific enzymatic tools for a definitive sequence analysis of this heteropolymer. Work in this direction is currently in progress in this laboratory.

SUMMARY

An alginic acid degrading enzyme, alginase, has been partially purified from cell-free extracts of an adapted pseudomonad and a study of its properties undertaken. The metabolism of al- ginic acid has been shown to involve the intermediate formation

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

316 Alginic Acid Metabolism Vol. 237, No. 2

of a series of unsaturated oligosaccharides and the resultant accumulation of a monosaccharide end product which has been isolated and characterized as 4-deoxy-L-erythro-5-hexoseulose

uranic acid.

REFERENCES

1. FISHER, F. G., AND DORFEL, H., Z. physiol. Chem., 303, 136 (1955).

2. WHISTLER, R. L., AND KIRBY, K. W., Z. physiol. Chem., 314, 46 (1959).

3. PREISS, J., AND ASHWELL, G., Federation Proc., 20, 86-D (1961).

4. PREISS, J., AND ASHWELL, G., d. BioZ. Chem., 237, 317 (1962). 5. ANET, E. F. L. G., AND REYNOLDS, T. M., Nature, 174, 903

(1954). 6. WARREN, L., Nature, 186, 237 (1960). 7. DISCHE, Z., J. Biol. Chem., 204, 983 (1953). 8. WEISSBACH, A., AND HURWITZ, J., J. Biol. Chem., 234, 705

(1959). 9. NAKADA, H. I., WOLFE, J. B., KOCHSTEIN, L. I., AND ANDREOLI,

A. J., Anal. Biochem., 2, 168 (1960).

10. SMILEY, J. D., AND ASHWELL, G., J. Biol. Chem., 236, 357 (1961).

11. MACFADYEN, D. A., J. Biol. Chem., 168, 107 (1945). 12. UMBARGER, H. E., AND MAQASANIK, B., J. Am. Chem. Sot.,

74, 4253 (1952). 13. LANNING, M. C., AND COHEN, S. S., J. Biol. Chem., 189, 109

(1951). 14. MEISTER, A., J. Biol. Chem., 197, 309 (1952). 15. LEVIN, D. H., AND RACKER, E., J. Biol. Chem., 234,2532 (1959). 16. LIPMANN, F., AND TUTTLE, J., J. BioZ. Chem., 169, 21 (1945). 17. SOWDEN,‘J. c., J. Am. Chem. koe., 76, 3541 (1954). 18. STUMPF. P. K.. J. Biol. Chem.. 169. 367 (1947). 19. ASHWELL, G., in S. P. COLOW&K AND N.‘O. KAPLAN (Editors),

Methods in enzymology, Vol. III, Academic Press, Inc., New York, 1957, p. 73.

20. WARAVDEKAR, V. S.. AND SASLAW, L. D.. J. Biol. Chem., 234, 1945 (1959).’ ’

21. LEVENE, P. A., J. Biol. Chem., 86, 785 (1930). 22. SUZUKI, S., J. Biol. Chem.. 235.3580 (1960). 23. LINKE$, A:, MEYER, K., A& HOFFMAN, P.; J. Biol. Chem., 219,

13 (1956). 24. LINKER, A., HOFFMAN, P., MEYER, K., SAMPSON, P., AND

KORN, E. D., J. Biol. Chem., 236, 3061 (1961).

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from

Jack Preiss and Gilbert Ashwell4-DEOXY-l-ERYTHRO-5-HEXOSEULOSE URONIC ACID

UNSATURATED OLIGOSACCHARIDES AND Alginic Acid Metabolism in Bacteria: I. ENZYMATIC FORMATION OF

1962, 237:309-316.J. Biol. Chem. 

  http://www.jbc.org/content/237/2/309.citation

Access 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/237/2/309.citation.full.html#ref-list-1

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

by guest on Novem

ber 22, 2018http://w

ww

.jbc.org/D

ownloaded from