Catalytic and structural properties of alginate lyases from bacterial epiphytes of Sargassum (Phaeophyceae)


Material Information

Catalytic and structural properties of alginate lyases from bacterial epiphytes of Sargassum (Phaeophyceae)
Physical Description:
vii, 151 leaves : ill. ; 28 cm.
Romeo, Tony, 1956-
Publication Date:


Subjects / Keywords:
Sargassum   ( lcsh )
Phaeophyceae   ( lcsh )
Energy crops   ( lcsh )
Lyases   ( lcsh )
Microbiology and Cell Science thesis Ph. D
Dissertations, Academic -- Microbiology and Cell Science -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1986.
Bibliography: leaves 140-150.
Statement of Responsibility:
by Tony Romeo.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000895871
oclc - 15376388
notis - AEK4476
sobekcm - AA00004869_00001
System ID:

Full Text








I offer my sincere gratitude to Dr. James Preston who

has provided immeasurable encouragement, support, and

guidance to me throughout my graduate studies. The members

of my graduate committee have offered a great deal of

helpful advice for which I am grateful: Dr. L. Ingram, Dr.

W. Gurley, Dr. H. Aldrich, and Dr. P. McGuire. I also wish

to thank Dr. J. Gander for valuable advice and comment on

parts of this work. The students in Dr. Preston's lab have

been good friends and helpful colleagues to me. I wish them

success and happiness.

The skilled efforts of Donna Huseman in preparing the

figures for this work and Adele Koehler in typing the

manuscript are greatly appreciated.

Jeanette Reinhardt provided electron microscopy of

.Sargassum protoplasts. B. Parten and Dr. B. Dunn carried

out amino acid analyses and N-terminal sequence analysis of

alginate lyase. The NMR analyses of substrates were

obtained by Sandra Bonetti and Cynthia Jackson. The amino

acid sequence similarity search was conducted by Dr. Michael

Little in the Dept. of Biochemistry of the- University of

Arizona. J.C. Bromley, D.R. Preston, and J. Beiswanger

provided assistance in culturing and harvesting bacteria for

enzyme isolations. All of these contributions have been

valuable to my work.

I would like to thank my beloved wife, Lori, for her

patience and moral support.



ACKNOWLEDGEMENTS....... ...................... .............. ii

ABSTRACT......... .............................................. vi


I OBJECTIVES AND BACKGROUND...................... 1

Objectives and Rationale........................ 1
Background...... ........................ ......... 3


Introduction. .................................. 18
Isolation, Properties, and Growth of Alginate
Lyase Producing Bacteria...................... 19
Intracellular and Extracellular Alginate
Lyases: Substrate Specificities and
Cleavage Patterns ............................ 24
Digestion of Alginate Present in Sargassum
Tissues by Alginate Lyases.................... 30
Preparation of Protoplasts by Mechanical
Disruption of Tissue...................... ...... 37
Discussion ...................................... 39

BACTERIUM.................................... 46

Introduction...................................... 46
Experimental..... ............................... 48
Results and Discussion............................. 56



Introduction ........................ ......... 78
Materials and Methods.......................... 79
Results.................. .................. ......... 86
Discussion........ ............................... 104

ON THE REACTION RATE.......... ....... ......... 111

Introduction ........................... .... ...... 111
Materials and Methods.......................... 113
Results........................................ 116
Discussion...................................... 128

VI CONCLUSIONS................................... 138

REFERENCES.......................... ................... 140

BIOGRAPHICAL SKETCH...................... 151


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Tony Romeo

August 1986

Chairman: James F. Preston, III
Major Department: Microbiology and Cell Science

Enzymes which catalyze depolymerization of alginate,

the major cell wall polymer of brown algae (Phaeophyceae)

were isolated and characterized from bacteria which were

part of the epiphytic flora of healthy tissues of Sargassum

species. Both Gram negative obligate aerobes and

facultative anaerobes produced activities with a range of

substrate specificities. Extracellular preparations of the

bacteria were highly endolytic while cell extracts were more

exolytic in their overall mechanisms. The activities were

capable of extensive degradation of the alginate present in

tissues of Sargassum.

A method for analysis of the depolymerization of (1-4)-

B-D-mannuronan (poly(ManA)) by high performance liquid

chromatography (HPLC) demonstrated that the single

extracellular alginate lyase from a fermentative isolate

generated unsaturated oligouronides which included dimeric

through pentameric products. The trimer was the major

product and could not be further depolymerized; the tetramer

was converted to trimer at a low rate; the pentamer was

readily converted to trimer and dimer forms. The

intracellular preparation differed in that it generated a

considerable amount of monomer.

The extracellular alginate lyase was purified to

chromatographic and electrophoretic (SDS-PAGE) homogeneity

by gel filtration on Sephadex G-75, anion exchange HPLC on a

Mono Q HR 5/5 column, and gel filtration HPLC on an Ultrapac

TSK-G4000SW column. The enzyme was composed of a single

polypeptide of 29 kDa. The amino terminal sequence was

determined through the first 6 amino acids, and 19 of the

first 30 amino acids were assigned. Several closely

migrating forms were separated by isoelectric focusing,

which have pi values ranging from 4.2 to 5.0, suggesting

posttranslational modification. The secondary structure was

approximately 74% a-helix by CD spectroscopy.

The extent of depolymerization of well characterized

alginates and block regions of alginate by the purified

enzyme was strongly correlated with the frequency of

mannuronic acid triad in the polymers and not to the

mannuronic acid diad frequency or mannuronic acid content,

suggesting a minimal recognition site of three sequential

residues. Unsaturated trimer and tetramer end products of

the reaction did not show appreciable inhibition of

activity. A model for the specificity of the active site of

the enzyme is presented.




Objectives and Rationale

The general objective of these studies was to identify

and characterize enzymatic systems capable of depolymerizing

the major cell wall polymer of the brown algae (Phaeo-

phyceae), alginate. A major impetus for the work has been a

need for methods which would allow members of this group of

organisms, in particular, species of the genus Sargassum, to

be readily cultured. A long range goal for this project has

been the development of systems of introduction of new

genetic material into the Sargassum species to alter their

growth characteristics and/or biochemical properties for

improvement of their biomass potential. Identification of

enzymes which specifically degrade the carbohydrate polymers

present in cell walls of the brown algae should allow

methods for producing protoplasts of the algae to be

developed. As has been the case with work on the higher

plants, this capability should simplify procedures for

culturing these organisms and allow the development methods

for the eventual genetic manipulation of Sargassum species.

The isolation and thorough characterization of enzymes

capable of catalyzing depolymerization of alginate will also



provide specific tools for analysis of brown algal cell wall

structure and biosynthesis and for modification and analysis

of alginate structure. Fermentative bacterial isolates

which degrade alginate may prove useful in improving the

rate and/or extent of conversion of brown algal tissue to

methane during anaerobic digestion.

The specific objectives of this project were to

1) Isolate bacteria from Sargassum tissue, which are capable

of growth upon alginate as a sole carbon and energy source,

and evaluate the secreted and the intracellular enzymatic

activities which catalyze depolymerization of alginate.

This included measurements of relative levels of enzyme

activities produced by each isolate, determination of

substrate specificities and general mechanisms, i.e., endo-

or exolytic nature of the activities which depolymerize

alginate, and comparisons of the abilities of various

preparations to degrade Sargassum tissues. 2) Develop

methods which permit the depolymerization of alginate to be

quantitatively and conveniently analyzed, including methods

for identification and quantification of specific products,

determination of the rates of product accumulation and/or

depletion, and determination of the limit products of a

particular depolymerization reaction. 3) Isolate and

purify at least one of the bacterial alginate lyases to

homogeneity and characterize its subunit composition,

primary and secondary structure, and its mechanism of

substrate depolymerization.



Biomass Potential of Phaeophyceae

The marine brown algae inhabit oceanic coastal waters

throughout the world. Although several species have been

commercially exploited for their anionic carbohydrates, in

particular the alginates, the bulk of the world's supply

remains untapped. The giant kelp Macrocystis pyrifera has

been evaluated for its bioconversion to methane and shown to

provide methane yields that are competitive with other

biomass and waste sources (Chynoweth et al., 1981). Studies

are in progress to determine the feasibility of farming

Macrocystic pyrifera (Neushul, 1977) and Laminaria

saccharina (Brinkhuis et al., 1984) for the production of

feedstocks for methane generation. Species of the genus

Sargassum represent an alternative source which includes

benthic species common to the colder waters and pelagic as

well as benthic species found along the coast of Florida and

other subtropical and warmer temperate waters. Trawling

collections of the pelagic species, S. natans and S.

fluitans, have placed their estimated biomass in the

Sargasso sea alone at 4 to 40 million metric tons (Parr,

1939). The high carbohydrate content of these algae and

their present lack of commercial exploitation make them

attractive as a potential source of biomass for conversion

to methane.

Carbohydrate Polymers of Phaeophyceae

The brown algae produce large amounts of carbohydrates,

e.g., almost 50% of the dry weight of the giant kelp,

Macrocystis pyrifera (Chynoweth et al., 1981). The

functions of the major carbohydrates are, in general,

related to the roles in maintenance of structural integrity

or in supplying short or long term energy reserves for the

plants. The present discussion will focus upon the

structural carbohydrates which are localized in the cell

walls of the organisms and are depicted in Fig. 1-1.

The major wall component is alginate, a linear 1-4

polymer of f-D-mannuronic acid and a-L-guluronic acid which

generally comprises 10 to 25% of the dry weight of brown

algae (Table 1-1). Alginate is a compound of considerable

commercial importance; over 8,000 tons are utilized annually

in the United States (Wells, 1977). The commercial value of

alginate has been partially responsible for generating

interest in alginate; a relatively large body of information

has been obtained regarding its fine structure and solution


Studies of Haug and coworkers have established that the

two constituent uronic acids of alginate are arranged into

homo- and heteropolymeric block regions of DP (degree of

polymerization) around 20 for the homopolymers, which are

interspersed in native alginate (Haug et al., 1966; Haug et

al., 1967). These polymers were obtained in relatively pure

form by mild acid hydrolysis of alginate, followed by




H H H2
Alginic Acid

R \ 0 O H lH HK H H


CH3 0O-R,

H 0

H 0
H 03SO 0
H 0 -0 H
SOH HiH R5 Fucoidin

Figure 1-1.

Chemical structures of the three polymers
which comprise the bulk of marine brown algal
cell walls (Phaeophyceae). Unit saccharides,
glycan bonds, and modifications of the
saccharides are indicated. Configurational
aspects of structures are not implied.


Table 1-1. Alginate levels in Phaeophyta species.


Alginic acid % dry wt.

S. fluitans

S. natans

S. filipendula

S. polyceratium

S. vulgare

L. cloustoni

M. pyrifera






a,b,c,dvalues are from Aponte de Otaola et al. (1983), Davis
(1950), Black (1950), and Chynoweth et al. (1981).



fractionations based upon differential acid solubil-ity of

the individual polymers. Isolated mannuronan regions,

poly(ManA), and guluronan regions, poly(GulA), have been

subjected to x-ray crystallographic studies which indicated

that the configuration the poly(ManA) resembles a flattened

ribbon, whereas that of the poly(GulA) forms a buckled chain

(Atkins et al., 1971). Nuclear magnetic resonance analyses

of alginate (Penman and Sanderson, 1972) suggest the Cl

conformation for the a-linked mannuronic acid residues,

indicating diequiatorial linkages to adjacent residues, and

the 1C conformation for the a-linked guluronic residues,

which would form diaxial linkages, further supporting the

polymer conformations which were based upon x-ray crystal-


The conformations of the block regions of alginate and

their relative abundance in the native polymer have

important effects on the solution properties and biological

functions of alginate. Only the poly(GulA) regions of

alginate bind to calcium and certain other divalent ions

with high affinity, thus rendering the purified poly(GulA)

insoluble or causing the native polymer to form a gel

(Smidsrod and Haug, 1965; Haug and Smidsrod, 1965a; Haug and

Smidsrod, 1965b; Kohn et al., 1968). Circular dichroism

studies of the interaction of calcium with poly(GulA)

sequences suggest that calcium mediates cooperative

interactions of regions containing at least 20 sequential

guluronic acid residues and thereby allows stable interchain


dimerization to occur. The investigators of this process

have developed a model to describe the calcium guluronate

complexes which has been dubbed the "egg box model" (Grant

et al., 1973; Morris et al., 1978; Rees et al., 1982). A

gel or three dimensional lattice is formed by interaction of

native alginate with calcium ions, wherein insoluble calcium

guluronate complexes are flanked by the soluble poly(ManA)

regions and heteropolymeric poly(ManA, GulA) regions. Since

the brown algae in almost all instances are marine

organisms, alginate is present in the cell walls as a gel

containing a mixture of metal ions, although due to the high

selectivity of alginate for calcium, and the relatively high

concentration of calcium ions in seawater; this is probably

the most prevalent ion (Percival and McDowell, 1967).

The relative amount of poly(GulA) present in alginate

will affect its gelling properties, and this can vary

depending upon the source of the alginate (Penman and

Sanderson, 1972; Haug et al., 1974). Some species of brown

algae have been shown to form tough, firm, holdfast tissues

using alginate with high levels of guluronic acid and in the

same plant produce flexible apical tissues of alginate

containing high levels of mannuronic acid (Haug et al.,

1974; Andresen et al., 1977).

The algae are apparently capable of increasing the

poly(GulA) content of the alginate in a given part of the

plant as it ages (Haug et al., 1974). This might be

accomplished by either degrading the alginate originally

present and replacing it with new polymers of different

structure or by altering the uronic acid composition of the

polymer which is already present. Laminaria digitata has

been shown to contain an enzyme which depolymerizes alginate

(Madgwick et al., 1973a) allowing for the former

possibility; the presence of an epimerase capable of

converting mannuronic acid residues of alginate to guluronic

acid has been detected in Pelvetia canaliculata (Madgwick et

al., 1973b), allowing for the latter mechanism. A

combination of these two means of altering alginate

composition may also occur. Interestingly, a mannuronan-

epimerase from Azotobacter vinelandii has been purified, and

it has been shown to be capable of altering the distribution

of the guluronic acid residues which it generates, in

response to calcium ion concentration (Skajak-Braek and

Larsen, 1985). Perhaps the activities of algal epimerases

are also responsive to conditions which may require the

formation of alginates with varying structural and

functional properties.

A polymer which is quite similar to alginate of the

brown algae is produced by certain bacteria, including

Azotobacter vinelandii (Larsen and Haug, 1971) and certain

isolates of Pseudomonas aeruginosa (Evans and Linker, 1973).

These bacterial alginates differ from the algal polymer in

containing acetylations of certain hydroxyl groups. The

isolates of P. aeruginosa are of interest in that alginate

producing organisms are frequently isolated from patients

with cystic fibrosis, where the organisms and the alginate


which they produce are believed to increase the morbidity of

the disease (Hoiby et al., 1977). Studies of the immune

response of cystic fibrosis patients to alginate (Bryan et

al., 1983; Speert et al., 1984) and the genetics of alginate

biosynthesis by P. aeruginosa (Roehl et al., 1983; Banerjee

et al., 1985; Goldberg and Ohman, 1984) are areas of recent

interest which should increase our understanding of the

chemical properties and biosynthesis of alginate.

Fucoidin or fucoidan is a sulfated fucose polymer,

which is found in levels from less than 1% of the dry tissue

weight for species such as Macrocystis pyrifera (Chynoweth

et al., 1981) to 24% for species such as Fucus spirilis and

Pelvetia canaliculata which are extensively exposed to air

during growth (Percival and McDowell, 1967). It consists of

a,l -> 2 linked L-fucose with sulfate esterified primarily

at position 4 (Fig. 1-1). The molecule is probably branched

at positions 3 and 4 and as isolated, may contain other

saccharides, including galactose, xylose, and uronic acids,

and metal ions as well (Percival and McDowell, 1967).

Fucoidin is.believed to reside in extracellular mucilaginous

material and non-fibrillar portions of the wall, and as a

result of its hygroscopic properties, may protect the brown

algae against dehydration (Percival and McDowell, 1967;

McCully, 1966; Evans et al., 1973).

Cellulose is a 8,1 -> 4 unbranched glucan which is an

important structural component of higher plant cell walls

and is consistently found in the cell wall of brown algae,


in levels which range from about 1 to 10% of the dry weight

of the plant (Percival and McDowell, 1967). Studies on the

histology of the walls of Fucus have identified crystalline

components which consist at least in part of cellulose

(McCully, 1970). Studies on the zygotes of Fucus indicate

that the shape is maintained to a large extent by cellulose.

Procedures which extract the other components, including

alginate, leave an intact sack-like structure which retains

the original form of the zygote (Quatrano and Stevens, 1976).

Enzymatic Depolymerization of Alginate

Alginate has been shown to be chemically depolymerized

by acid hydrolysis, base-catalyzed s-elimination, and by a

free radical mechanism (Haug et al., 1963; Smidsrod et al.,

1963; Smidsrod et al., 1965). Although enzyme activities

capable of depolymerizing alginate were reported as early as

1934 by Waksman et al., the first definitive study of

alginate depolymerization which identified the s-elimination

reaction as the mechanism of bond cleavage was presented by

Preiss and Ashwell in 1962. This established that the

enzyme(s) responsible was a lyase, as opposed to a

hydrolase. Since that time virtually all alginolytic

enzymes examined, with one exception (Stevens and Levin,

1976a; Stevens and Levin, 1976b), have proven to be lyases.

This type of reaction also has been observed for enzymes

which depolymerize other uronic acid containing polymers,

including pectin and pectic acid (Collmer et al., 1982;


Fogarty and Kelly, 1983), heparin (Yang et al., 1984),

chondroitin (Linn et al., 1983), the acidic heteropoly-

saccharides of Rhizobium trifoli (Hollingsworth et al.,

1984), and many others.

The elimination reaction catalyzed by these lyases

relies upon the electron withdrawing carboxyl group of the

substrate (C-6), an extractable a-proton (at C-5), and an 0-

linked uronide (at C-4) which is the leaving group (Kiss,

1974). In the case of endolytic cleavage (internal to the

polymer), the new reducing terminus of the products is

identical to that which would be produced hydrolytically,

and the new nonreducing terminus is a 4,5 unsaturated

residue (see Fig. 1-2).

Assays for bond scission by alginate lyases measure

either generation of reducing or nonreducing termini.

Unsaturated termini absorb UV with a maximum at

approximately 232 nm (Preiss and Ashwell, 1962a), although

the unsaturated monomer, 4-deoxy-L-erythro-5-hexoseulose

uronic acid does not do so appreciably. A specific and

sensitive assay for both unsaturated monomer and unsaturated

nonreducing terminal residues is based upon the spectrophoto-

metric determination at 548 nm of the chromogen generated

upon the reaction of 2-thiobarbituric acid (TBA) with

periodate treated products (Preiss and Ashwell, 1962a).

Alginate lyases have been shown to be produced by and

in some cases isolated and characterized from marine

invertebrates (Nakada and Sweeny, 1967; Favorov and





co- Coo

0 0



Alginate lyase



coo- Lo

Figure 1-2.

The alginate lyase reaction as catalyzed by
an endo-poly(ManA) lyase. A new reducing end,
indistinguishable from that formed by
hydrolysis, and a 4,5 unsaturated nonreducing
end which confers UV absorbance properties
and reactivity of the lyase products in the
TBA assay are the result. The configurations
of the poly(GulA) and poly(ManA) regions of
alginate are indicated.



Gul Man





Vaskovsky, 1971; Elyakova and Favorov, 1974; Favorov et al.,

1979; Muramatsu et al., 1977; Muramatsu, 1984; Jacober et

al., 1980), fungi (Wainwright and Sherbrock-Cox, 1981),

marine bacteria (Kashiwabara et al., 1969; Fujibayashi et

al., 1970; Min et al., 1977; Davidson et al., 1976; Quatrano

and Caldwell, 1978; Doubet and Quatrano, 1982; Doubet and

Quatrano, 1984; Pitt and Raisbeck, 1978; Southerland and

Keene, 1981; Preston et al., 1985a), terrestrial bacteria

(Boyd and Turvey, 1977; Boyd and Turvey, 1978; Hansen et

al., 1984), and brown algae (Madgwick et al., 1973).

The work of Nakada and Sweeny (1967) initially

demonstrated that an alginate lyase from abalone showed a

preference for alginate containing a high mannuronic acid

content, and a second enzyme from the same source was more

active with alginates that were high in guluronic acid. All

alginate lyases which have been subsequently examined are

selective in their substrate specificities; i.e., no enzyme

has been isolated which shows significant activity upon both

poly(ManA) and poly(GulA). However, there are reports of

two lyases which will cleave both the GulA-GulA bonds of

poly(GulA) and the GulA-ManA bonds of poly(ManA, GulA) (Boyd

and Turvey, 1978; Min et al., 1977).

Alginate lyases may attack their substrates by either

endolytic or exolytic mechanisms, keeping in mind that some

polyuronide degrading enzymes act by making apparent random

initial endolytic cleavages followed by non-random cleavages

in the later stages of substrate depolymerization (Thibault,


1983). The mechanism of substrate attack by alginate lyases

was first examined by viscometric analyses (Nakada and

Sweeny, 1967) and thence has also been examined by analysis

of reaction products, using methods which have included

conventional gel filtration and ion exchange column

chromatography (Boyd and Turvey, 1977; Favorov et al.,

1979), paper electrophoresis (Davidson et al., 1977), and

gel electrophoresis (Doubet and Quatrano, 1984).

The rigorous examination of the catalytic activities

and structural properties of an enzyme require its

purification and characterization at least to an extent

which establishes that a single activity has been obtained.

Preferably, the enzyme will have been purified to a

single homogeneous protein. Alginate lyase enzymes have,

however, in most cases been analyzed in impure states.

Davidson et al. (1976) reported that they had purified a

poly(GulA) lyase from a marine bacterium, although the data

which were presented did not establish their claims.

Enzymes from marine molluscs have been purified, all of

which were poly(ManA) lyases (Elyakova and Favorov, 1974;

Muramatsu et al., 1977). Recently, a bacterial poly(GulA)

lyase was purified to electrophoretic homogeneity (Doubet

and Quatrano, 1984). Studies of Muramatsu and coworkers on

two isoenzymes from the snail Turbo cornutus represent the

only thorough studies of the structural properties of

alginate lyases up to this time (Muramatsu and Egawa, 1982;

Muramatsu et al., 1984). These two enzymes were each

composed of a single subunit of 32,000 kDa by SDS


polyacrylamide gel electrophoresis. Native masses of 25 kDa

were established by Sephadex G-100 chromatography. The

isoelectric points were 7.5 and 7.7 for SP1 and SP2 enzymes,

respectively. The enzymes were glycoproteins and were

composed primarily of B-sheet secondary structure.

Enzymatic Digestion of Brown Algal Cell Walls

The presence of cellulose in the cell wall of the brown

algal wall suggests a requirement for cellulases, in

addition to alginate lyases specific for each of the block

regions of alginate, to effect complete digestion of wall

material. Quatrano (1982) reported that the cellulose which

is deposited during early development of Fucus is subject to

depolymerization by cellulases. The cellulases have been

extensively studied (Enari, 1983) and are readily available

from a variety of commercial sources.

Although the fucoidin of the cell wall may not

contribute to the shape or structural base of brown algal

tissue, it may pose a barrier to cellulases and alginases

and thereby prevent or impede the removal of the cell wall.

There are reports of bacteria having been isolated which

produce enzymes capable of depolymerizing fucoidin (Quatrano

and Caldwell, 1978; Morinaga et al., 1981; Yaphe and Morgan,

1959). However, no such enzymes have been isolated and

examined in detail.

There is a recent report of a method for enzymatically

removing the cell walls of Sargassum species to generate


protoplasts by the use of extracts of the hepatopancreas of

abalone (Preston et al., 1985b). The method is, however,

not yet reproducible, and the origins) of the difficulties

are at this time unknown. A reasonable approach to

developing reliable methods for obtaining protoplasts from

Sargassum species might be to fractionate and reconstitute

extracts of abalone. Individual enzymatic factors necessary

for generation of protoplasts might then be identified and

isolated more reproducibly from bacterial and/or fungal





Certain members of the brown algae, or Phaeophyceae,

have been exploited for their anionic carbohydrates,

specifically alginate, for many years (Percival and

McDowell, 1967; Steiner and McNeely, 1954). However, the

following studies were prompted by the more recent interest

in their potential as a source of biomass for conversion to

methane (Preston et al., 1985b). Species of Sargassum are

particularly attractive in this regard, as they are not

commercially exploited at present; i.e., there are no

competitive uses for them, and they contain significant

quantities of the carbohydrates alginate (Aponte de Otaola

et al., 1983) and mannitol (Preston and Jiminez, 1986). The

ability to reproducibly generate protoplasts from Sargassum

species would aid efforts to develop methods for tissue

culture and for genetic improvement of these algae (Preston

et al., 1985b).

The goal of the work described in this chapter has been

to explore potential enzyme systems for degrading the cell



walls of Sargassum. In particular, alginate lyases, which

depolymerize the major wall polymer, alginate, have been

assessed for their activities on Sargassum cell walls and

for their capacities to generate protoplasts of Sargassum


The natural bacterial flora of Sargassum have provided

the source of the alginate degrading enzymes for this study.

The substrate specificities and overall mechanisms of the

intracellular, or cell-bound enzymes, and of the secreted,

or extracellular enzymes, of the bacteria have been

examined. The effects of these activities on algal tissue

have been quantified by a sensitive and specific assay for

the unsaturated nonreducing terminal residues which are

generated by the eliminative cleavage of alginate by

alginate lyases (Preiss and Ashwell, 1962a). This has

allowed comparisons to be made of the efficacies of

individual alginate lyase preparations, which differ in

substrate specificities and mechanisms, to remove alginate

from Sargassum cell walls.

Isolation, Properties, and Growth of Alginate
Lyase Producing Bacteria

Viable tissues of Sargassum natans and Sargassum

fluitans provided the source for isolation of alginate lyase

producing bacteria, except in the case of isolate FM, which

was obtained from decaying Sargassum. The algae were

transported from the Atlantic Coast of Florida to


Gainesville, Florida, where identifications were confirmed,

voucher specimens saved, and epiphytic bacteria isolated.

Sargassum tissue (1 g quantities) was subjected to mild

sonication in sterile sea water (Instant Ocean from Aquarium

Systems, Mentor, OH) to dislodge epiphytic bacteria.

Dilutions of the sea water were plated onto solid alginate

medium (2% agar, 1% sodium alginate in PESI, Provasoli's

enriched seawater, supplemented with 0.27 g/L iodine)

(Provasoli, 1968; Polne-Fuller et al., 1984). Colonies

which exhibited substantial clearing of the calcium alginate

haze of the medium after several days of growth were

selected for further studies. Purity of the cultures has

been established by subculturing on solid alginate medium,

growth in liquid alginate medium (0.1% sodium alginate, PESI

containing 1.0 mM calcium and 5.5 mM magnesium), and growth

on a rich solid medium (2% agar, 1% glucose, 0.8% nutrient

broth, 1% yeast extract in PESI).

Some of the morphological and physiological properties

of the bacteria have been described (Preston et al., 1985a)

and along with other properties are shown in Table 2-1. All

isolates are Gram negative, polarly flagellated rods. Of

seven organisms isolated, four are oxidative and three

fermentative. All oxidative organisms are oxidase positive

and all fermentative isolates oxidase negative. All

fermentative isolates, but none of the oxidative bacteria,

produce acid in liquid glucose medium (Table 2-1). None of

the isolates showed evidence of gas production on glucose or


*m r
* *
in O v














0 O
0 (

-0 0



aw CN

4- 0
O 1




a 0

a -

-4 (a








(C4 1 ICD- m
4N M CM4 M C4 1

0 .
--4 -4 N r-4 -4 -4

4-1 X X X( X xx

o -i o m o LAn L

OCLn r-


(N 00 00 00 0

S0000 0


+ I + + I + +1

(-1 -4
0) -r4

(U C

C 41

0 V
w a)
4-4 N

C 3


r1 >





s4 a

0 a


C j

-0 m)
0 *0
o 3

04) W W
0 4













a o












-4 *


**. m 1-4
CO ^D <
*o *o
in 4a 1s


x r





0 3


-4 r



w -4


0 14

^ -4












C -4


- 0



alginate containing media. Morphological, physiological,

and DNA base composition data allowed the assignment of

aerobes to the genus Alteromonas (Preston et al., 1985a).

The fermentative isolates, although morphologically and

physiologically similar to bacteria of the genus

Photobacterium, are excluded from this genus by their DNA

base composition and have not been assigned to any existing


Figure 2-1 shows the growth of two representative

fermentative isolates, A (SFFB080483 A) and G (SFFB080483

B), in liquid alginate medium at 220C with rapid gyrotory

shaking. Duplicate flasks of media were periodically

sampled, the turbidity was measured, and cells were removed

by centrifugation prior to measurements of alginate

utilization and generation of alginate degradation products.

Isolates A and G removed alginate from the medium during

growth, as did all of the isolates selected for these

studies (data not shown) as measured by a loss in total

uronic acid equivalents from the medium (Blumenkrantz and

Asboe-Hansen, 1973). The depolymerization of alginate to

form oligomeric uronides which possessed unsaturated

nonreducing terminal residues, as measured by the method of

Preiss and Ashwell (1962a), occurred during the growth of

the isolates. This indicated that alginate was being

degraded by enzymes which were transeliminases or lyases.







0.8 q


0 12 24
Time, hours

Figure 2-1. Growth of bacterial isolates in liquid alginate
medium. Isolates A (a) and G (b) were cul-
tured as described in the text. Growth was
monitored by measuring turbidity (A600), total
uronic acid equivalents (A520) ad unsaturated
nonreducing terminal residues (A 48).

0.2 -

0.4- b


For enzyme isolation, bacteria were grown in Fernbach

flasks containing 1 1 of liquid alginate medium at room

temperature (220C) with rapid gyrotory shaking and were

harvested at late exponential phase.

Intracellular and Extracellular Alginate Lyases:
Substrate Specificities and Cleavage Patterns

Isolation of Bacterial Enzymes

Bacterial cells were removed from culture medium by

centrifugation, frozen in liquid nitrogen or at -700C, and

stored at -700C. For analysis of extracellular enzymes the

spent medium was concentrated by tangential flow filtration

using a Millipore Pellicon cassette system with a

polysulfone (PTGC) membrane which allowed retention of

molecules larger than 10 kDa, and dialyzed against

distilled, deionized water.

For intracellular preparations cells were thawed,

suspended in 4 volumes of ice cold 0.1 M sodium phosphate

buffered at pH 7.5, and disrupted with a French pressure

cell at 16,000 lb in-2. Unbroken cells and cell debris were

removed by centrifugation at 10,000 x g for 15 min, and

acidic polymers were rendered insoluble in the supernatant

solution by adding 5% streptomycin sulfate dropwise with

stirring to a beaker at OOC to give a final concentration of

2%. After stirring the mixture for 10 minutes at 00C, the

resulting precipitate was removed by centrifugation and the

supernatant solution containing alginate lyase was treated


with solid ammonium sulfate (to 65% saturation). The

protein precipitate was pelleted by centrifugation at 10,000

x g, 10 min, 40C, redissolved in pH 7.5 phosphate buffer,

and dialyzed against distilled deionized water or sodium

phosphate buffered at the desired pH.

Preparation of Substrates

Sodium alginate was purchased from Fisher Scientific

Company as a purified grade originally isolated from

Macrocystis. Prior to use in viscometric determinations

alginate was centrifuged at 100,000 x g for 5 h (Smidsrod

and Haug, 1968). Poly(ManA) and poly(GulA) were obtained

from HC1 hydrolyzed alginate, following the methods

developed by Haug et al. (1967). Preparations of poly(GulA)

and poly(ManA) were further fractionated on Sephadex G-50

with 0.5 M NaCl as eluant, and selected fractions analyzed

by reducing sugar and total carbohydrate assays (Nelson,

1944; Dubois et al., 1956; Haug and Larsen, 1962) and 1H and

13C NMR (Grasdalen et al., 1979; Grasdalen et al., 1981), to

assess uniformity of size and purity of substrates. The

poly(ManA) fraction contained approximately 89% B-D-

mannuronic acid residues and contained polymers with degree

of polymerization (DP) values of 16-20; the poly(GulA)

contained 89% a-L-guluronic acid residues and had an average

DP of 22.


Substrate Specificities of Intracellular and
Extracellular Preparations

Alginate lyase was quantified by the TBA assay (Preiss

and Ashwell, 1962a; Weissbach and Hurwitz, 1959). Substrate

mixtures contained either 0.1% sodium alginate, poly G, poly

M, or no carbohydrate (controls for endogenous substrate),

in 0.05 M KC1, buffered with 0.03 M sodium phosphate from pH

5 to 8, or 0.05 M sodium acetate at pH 4.

With three exceptions, activities of the enzyme

preparations with each substrate were maximal at pH 8.

Extracellular preparations from facultative organisms A and

D were most active on poly(GulA) at pH 7 and extracellular

activity of isolate G was highest with poly(ManA) at pH 7.

Figure 2-2 compares intracellular and extracellular

activities from isolate A on poly(GulA), poly(ManA), and

alginate under several pH conditions.

Comparison of levels of activities of intracellular and

extracellular preparations on poly(GulA), poly(ManA), and

alginate at pH 8.0 are shown in Table 2-2. Intracellular

preparations were, in all cases, most active on poly(ManA)

and generally slightly higher on alginate than poly(GulA).

Extracellular preparations generally were highly active on

alginate. Fermentative isolates A and D showed little or no

activity toward poly(GulA) extracellularly, and fermentative

isolate G showed little activity on poly(ManA). Levels of

extracellular activities of the oxidative isolates, FM, B,

and C, were comparable on poly(GulA) and poly(ManA).


,poly G
4 6 8

lob 100 A Intracellular
100 poly M

*3 /
Iop %


4 6 8

Figure 2-2. Activities of intracellular and extracellular
preparations from isolate A, under various pH
conditions, toward poly(GulA), poly(ManA),
and alginate. Values have been normalized to
the condition which allowed the maximal number
of bonds to be cleaved, as determined using
the TBA assay.
'Ie Iodto hc loedtemxmlnm
*f Iod ob lae, sdtrie s
the~ TE Gsay


* 4 1

" i--

l lo























3 4-1



mr o moo

M ri oO

0000 0 -

"M -W C r r-4 -r

SP-41 00 o C

0 o( t Ln in r-
m m CN r-I CV CO
-I o0 0 CO C



0r r-4 rl -4

-00 o 00 r



-V4-4-4-4N -4-N


r- Ln r- co 1-1 mn
v D N mco o m

in m CN n 0 M

OOOO o m
m 0 C0 ',O o- 0N

o CD 0 0 o 4


4J En VQ)
a) >>

E 0 > 0
S0) 4J (0 -c

SC m C

wOM 0 -cC
04' 04,C Q)-

CC 0! 4 c w
Q) 0 O c
w- c wO ( 24
Cl aC w a

E >

C0 44
0w C 4 0 a
04 4J 0

-HC V O LH >

4J0 c *

S0 C 04.

0 O>) 44

W- -a) a 4J 1
4lO W M U ) Ca E
44 CV H a)
LO 0 a W- 4- j
H n L Wr '-W CI a

a) .4 Cr 04 -
C 0) u 4 i .w
--H = O
(ci4-4 U) -4 O ,
4J M EU ) -4 >9
4W 4-4- > '
o) ar Ho 4 -H C

0.' c 0

0 -c C r
4 14 a 04 E 1- 4

C 0 0 O 0

-4J J Mw : 4
0 W -1 -4 )-4 U) c U-r C ^ >

1-4I44 -HO 0 (o q
4) 4I 34-4. UC 4' 44 Oa

a0 C C .- (o 0 nl -44LWL
HHH 4a >"0 >) Oa0

>0l 0 >0 0-
E-4 3c a)

Patterns of Substrate Cleavage

Measurement of the relative level of endo- and

exoeliminase activities in intracellular versus

extracellular preparations was carried out at pH 8.0. The

decrease in the viscosity of alginate during

depolymerization was measured by capillary viscometry (McKie

and Brandts, 1982), and the rate of glycan bond cleavage was

determined by the TBA assay. Plots of the reciprocal of

specific viscosity, i.e., specific fluidity,
periodic acid generated TBA reactive products produced

straight lines with slopes proportional to the relative

level of endolytic activity. In the organisms examined, the

slopes were greatest, and therefore the endolytic activities

highest, in the extracellular fractions. The oxidative

isolate FM, in particular, shows striking partitioning of

exo- and endolytic activities. The comparisons of these

slopes are given in Table 2-2.

Digestion of Alginate Present in Sargassum Tissues
by Alginate Lyases

Digestion of S. filipendula Tissues by Intracellular
and Extracellular Alginate Lyase Activities from
Isolate A

Active apical tissue with no visible epiphytic growth

was excised and subjected to mild sonication, weighed,

finely chopped with a scalpel, and incubated with enzyme

preparations from facultative isolate A under conditions

described in Table 2-3. The total number of unsaturated


Table 2-3. Degradation of S. filipendula tissue by
intracellular and extracellular alginate lyase
preparations from isolate A.a

Direct Assayb +Enz. Mixturec
n.r. ends u moles n.r. ends u moles

h Intra. Extra. Intra. Extra.

0 0.005 0.004 -- --

2 0.131 0.049 0.140 0.163

6 0.317 0.167 0.371 0.519

10 0.419 0.226 0.432 0.703

24 0.651 0.334

aTwo 25 mg apical portions of sonicated S. filipendula
tissue, containing an estimated 2.5 u moles of uronic acid
residues each, were finely chopped and incubated with enzyme
preparations in PESI lacking added Ca ++ and Mg +, and
containing 1.2 mM EDTA. The activities of both enzyme
solutions at the start of the experiment were 0.0198 u
mole/min per ml with alginate as a substrate.

bSamples of sea water media were removed at indicated times
and assayed for products using the TBA assay, which measures
nonreducing (n.r.) unsaturated terminal residues.

cSamples of the seawater solutions (excluding tissue) were
removed at the indicated times and added to a mixture of
intra- and extracellular alginate lyase preparations
containing activities of 0.099 u mole/min per ml from each
source, incubated further for 12 h and assayed for products
of the lyase reactions as above.


nonreducing termini produced by the intracellular extract

was at each time point greater than that produced by the

extracellular extract. However, when samples of each

reaction mixture (minus tissue) were removed, mixed with a

combination of intracellular and extracellular enzymes, and

allowed to incubate further, the material released from the

tissue by the extracellular preparation was observed to be

accessible to further depolymerization, whereas the material

released by the intracellular preparation was not. The most

plausible explanation for this is that the extracellular

preparation, which was shown in Table 2-2 to be highly

endolytic, depolymerized the alginate of the tissue to yield

oligomers which were substrates for the enzymes in the

second incubation. The intracellular enzymes were also

capable of releasing and depolymerizing alginate from the

tissue, but released less total mass of alginate. The

material released by the intracellular preparation was in a

more highly depolymerized state than that released by the

extracellular preparation and could not be further degraded

by the enzyme mixture in the second incubation. Protoplasts

were not quantitatively released from the tissues in any of

the above digestions, and under light microscopy the tissue

appeared relatively intact.


Degradation of Tissue-Bound Alginate by Extracellular
Enzymes from Isolates A and G

In order to examine the capacities of extracellular

preparations which release unsaturated oligomeric products

from alginate in tissues of S. filipendula and S. fluitans,

the experiment described in Table 2-4 was carried out. The

preparation from isolate G, which was active primarily on

poly(GulA), generated approximately 4-fold as many

unsaturated termini from S. filipendula as did the enzyme

from isolate A, which is active with poly(ManA) but not

poly(GulA). When samples from individual wells were added

to the complementary enzyme preparations and incubated

further, the products generated by extensive degradation of

tissue with isolate G enzyme preparation (24 h) could not be

further depolymerized in the second incubation. The

products from 24 h degradation by isolate A enzyme

preparation were depolymerized to yield 2-fold more

unsaturated termini. The total wall mass released by the

preparation from isolate G was therefore 2-fold greater than

that released by the preparation from isolate A.

Both enzyme preparations released approximately the

same amount of wall mass from the S. fluitans tissue,

although the preparation from isolate G released 2-fold more

unsaturated termini. In addition, the tissue from S.

fluitans, unlike that from S. filipendula, was depolymerized

more effectively after treatment with EDTA.


CM CM in 10 CD a



n n 00 C 00

0 .0 N 0 0
0 0 CM N 0 0


r 0o CM o











4a En





0- r-I

a M

'U E

(0 '
rn .-
3 '-1

S u

. >4-
4-1 (

0 M

cu <

tr e
Q '






0 0 00 r-4 r-


) r n CV n 1

*r-* r- C 0 C C
* **q *I C

co w


--4 V

0 0

o C o -4


0 C)
M0 0


c c







rn c


o 1

-ro 0



C a)
E -W -W )
n 1-1 C a) a 0
(U (ocUU3 '4.

'4. >3 0D O >-4
S0 i10 O 30

z a2 Va)

6.'4 1 5 03
u MI <" 3C 4 .0 (u 3 v0 a
c ) gO 0 4OJ a) :o
0 c O 44 3 -1
O- c o" 0 E (U C
44 oC 0 c-u >

SCD u C) 0 a C4 V
C E % <. ( U ) O E
0 0^. >0 0

r0J (u 00 E- 0

C) C O ') 0
i210C -4o N ( oo.~u
)0 0 34 -C -4 C

C.0 1 4 C)

w W (U r 4-

0 a .0 a 00

.C C >i 3a 1 U' )
ro 0 c

C0 w a 3:
(a co ra x C: 4 0

t .- Q 0 m 0( ( 4-)0 NU -
U) r- mie
SM 1 4 0 Q)

V E 3 ja 0J 44 J > (
-1 c 0) M M

'-I C U C1 V C U C) C -W 4 x
Ui V ) MC U 3 (4 0 a w *q
C) *- 4 C 0 00-41 .4 4 >
-0 0 -0 M m C

C C 4-i 4 > -CI C 1 En -

O 0C)0 g U 4V C 00-!4! -
.C O -'O 4 C N3 0
ra r

0 C (1 S (D 4e C aa X
a 0 C)M C U 0 V C4.H

44 O Ui-H r
ra E 0 m 5
r a\ M 0 0 -W M

* id o oc 6 ue

C) -)> 4U) CD 't lt r u o D ) (o ><
3a se1 >4 w t a o

O4 0 C00 0 CO -U C U
LW 0 m 0 m w 0 c rn aW

.o 0 040 C C 44 4 OC O
0 i i m 30 i 0 En a 3 -c

( -O C U )- C 4 -C)

_U 0 0 M m -
4 4M .- O O 0 wN
0 :3 p a 4J : -4 -H -W

U aO ( a O O C C
00 4j C O: F-: 5 4

r- ( 0 W wE
0N -4 0 W U3 0i C c M. Q O 3

-4 w ,Q 0 m c


It is interesting that the level of products from S.

fluitans decreases between 8 and 24 h. In a similar

experiment, this was observed to occur only when tissues of

this species are degraded, since three other Sargassum

species, S. filipendula, S. furcatum, and S. hystrix did not

show similar decreases (data not shown). The tissues of S.

fluitans, from which the bacterial isolates were obtained,

may harbor more bacteria capable of utilizing the products

which accumulate. Alternatively, other species of algae may

produce inhibitory compounds which prevent bacterial growth

and subsequent utilization of depolymerized alginate.

Effect of Cellulase on Cell Wall Digestion
by Alginate Lyase

Cellulose which is present in the cell walls of brown

algae may present a barrier to release of protoplasts from

algal tissues and/or impede the digestion of alginate by

lyases. The effect of cellulase on the release of alginate

fragments by alginate lyase from S. fluitans was examined as

follows. Apical tissue was sonicated briefly to remove

epiphytes, and 100 mg samples were sectioned into small

pieces (< 0.5 mm in width). These were incubated for 10 h

in individual wells of a microtiter plate in artificial

seawater which lacked Ca++ and Mg++ salts and contained 50

mM EDTA. Seawater solutions were removed and fresh

solutions containing alginate lyase or alginate lyase plus

cellulase were added to the wells. The cellulase was


prepared from Trichoderma viride (purchased

from Sigma Chemical Co., desalted by chromatography on

Biogel P-6) and the alginate lyase was from an acetone

powder preparation of abalone entrails (from Sigma Chemical

Co., desalted by P-6 chromatography). Each preparation was

dissolved at a concentration of 20 mg/ml. At this

concentration 1 ml of the alginate lyase generated 0.55

umoles of product/min from sodium alginate and the cellulase

solution, as prepared, was sufficiently active to

quantitatively convert calli of cultured Daucus carota to

protoplasts in less than 30 min.

Figure 2-3 shows that the addition of cellulase did not

improve the conversion of cell wall bound alginate to

soluble unsaturated uronides by the alginate lyase

preparation, although this does not imply that cellulase is

not removing cellulose from the walls. This combination of

cellulase and alginate lyase enzymes was not effective in

quantitatively converting algal tissue to protoplasts.

Preparation of Protoplasts by Mechanical
Disruption of Tissue

During efforts to enzymatically remove the cell walls

of Sargassum sp., protoplasts were observed consistently to

be generated in low numbers (1000-2000 per 0.10 g of

tissue), although the bulk of the tissue remained intact.

These protoplasts were produced by the mechanical disruption






4.0 8.0
TIME (h)

Figure 2-3. Digestion of S. fluitans tissue with alginate
lyase in the presence and absence of cellulase.
Apical tissue was prepared from S. fluitans,
incubated in a microtiter plate with alginate
lyase from Haliotus (abalone) in the presence
(closed triangles) or absence of cellulase
(open circles) from Trichoderma viride,
and the contents of the wells sampled and
assayed for nonreducing termini by the TBA
method according to the text. Controls with
no added enzymes are indicated by open


of tissue, in the absence of any added enzymatic activity,

by slicing tissue into small pieces which were then

incubated in PESI solution which lacked Ca and Mg

Figure 2-4 shows what are apparently intact as well as

damaged protoplasts. The intact protoplasts also showed

neutral red staining of vacuoles (Stadelmann and Kinzel,

1972), trypan blue dye exclusion, and intense red

fluorescence of chloroplasts under UV light (data not

shown), which are indicative of viability (Berliner, 1981).

The protoplasts rapidly lysed upon exposure to distilled

water and had smooth outer surfaces as viewed by scanning

electron microscopy (Fig. 2-5), suggesting a membrane

surface which is free of cell wall. At least some of the

protoplasts of a given preparation remained viable for

several hours, although their capacity for regeneration or

continued growth was not tested.


This chapter and our previous work (Preston et al.,

1985a; Romeo et al., 1986) document observations that

fermentative as well as oxidative marine bacteria are

capable of producing alginate lyases of varied substrate

specificities. Marine bacterial poly(GulA) lyases were

previously described by several investigators (Fujibayashi

et al., 1970; Davidson et al., 1976; Sutherland and Keen,

1981) and oxidative bacteria from Fucus, which produced both

poly(GulA) and poly(ManA) lyases were isolated by Doubet and



19k ** <

*. lm *

Di S

Figure 2-4.

Light microscopy of protoplasts from S.
fluitans. Healthy apical tissues of S.
fluitans were sonicated briefly to remove
epiphytes, weighed (0.1 g), and sliced into
small pieces with a scalpel, and incubated
0.5 h in PESI lacking Ca and Mg++. The
suspension was drawn into a Pasteur pipette,
avoiding large pieces of tissue, transferred
to a clean well of a microtiter plate, and
observed at 400X, with an Olympus inverted


Figure 2-5. Scanning electron microscopy of protoplasts
from S. fluitans. Panel a depicts a single
protoplast emerging from algal tissue; panel
b shows two protoplasts resting upon a piece
of Sargassum tissue (1600X).


Quatrano (1982, 1984). Intracellular extracts from our

isolates were high in poly(ManA) lyase, making them similar

to the isolates from Fucus (Doubet and Quatrano, 1982).

Extracellular preparations, except from isolates A and D,

were most active on native alginate. Levels of poly(GulA)

and poly(ManA) lyases were comparable in the extracellular

preparations from all oxidative isolates that were examined.

On the other hand, fermentative isolate G produced much

more poly(GulA) lyase than poly(ManA) lyase, and isolates A

and D produced poly(ManA) lyase in large excess over

poly(GulA) lyase. These last two organisms are in contrast

with the isolates from Fucus which tended to produce higher

levels of poly(GulA) lyase in the extracellular fractions.

An organism such as isolate A, which produces little if any

extracellular poly(GulA) lyase but makes intracellular

poly(GulA) lyase may depend upon other bacteria in the

environment to secrete endo poly(GulA) lyases for the

further depolymerization of alginate.

The observation that endolytic activities are higher in

the extracellular fractions as compared to the intracellular

fraction presumably reflects a requirement for degradation

of large native alginate molecules to allow entry into the

bacterial cells. By retaining exoeliminases either in the

cytoplasm or bound to the cell surface, these organisms avoid

producing large amounts of metabolizable monomers in the

external environment which through diffusion and/or

utilization by other organisms would be lost to the bacteria

producing the enzymes.


A model for bacterial utilization of alginate based on

this study and the work of others is shown in Fig. 2-6.

Native alginate is endolytically depolymerized to fragments

possessing unsaturated nonreducing ends. These fragments

are internalized and exolytically degraded, possibly

undergoing degradation during the entry process. The

metabolic pathway for degradation of the monomer product by

a pseudomonad has been described (Preiss and Ashwell,


Under conditions of anaerobic digestion, the

metabolites of the monomer should be readily converted to

methane by a consortium of bacteria. Shiralipour et al.

(1984) have observed that improved yields of methane can be

obtained from digestion of Sargassum tissues using

microflora associated with Sargassum. The rate and/or

extent of methane production might be further improved by

supplementary inocula of organisms such as the facultative

isolates of this study, which produce alginate lyases with a

spectrum of substrate specificities and modes of cleavage,

or by addition of alginate lyase enzymes to the fermentor.

The exposure of active Sargassum tissues to bacterial

alginate lyase preparations is effective in removing

considerable amounts of alginate from the cell walls

(virtually all of the alginate may be removed from the

tissues by endolytic extracellular alginate lyase

preparations). Although alginate lyase activities alone or

in combination with a cellulase preparation did not effect



Endo Poly (GulA) Lyase A X-G
o AX-M
Endo Poly Unsaturated
o o (ManA) Lyase Fragments
o of Alginate



A X-M-M I O Further
A X-M HOCH Metabolism
Exolyases HCH
Uronic Acid

Figure 2-6. Proposed model for metabolism of alginate by
bacteria which colonize tissues of Phaeophyceae.
Alginate polymers are released from a gel-like
state in the wall by secreted endolytic bac-
terial alginate lyases. The initial products
are depolymerized to an extent which allows the
bacteria to assimilate them. Exolytic enzymes
continue the depolymerization of oligomers
which enter the cell, with the eventual produc-
tion of the monomer, which may be used as a
source of carbon and energy.


quantitative release of protoplasts from tissue, the

combined effects of these activities and activities toward

other cell wall polymers, e.g., fucoidin and/or proteins,

may be successful. The mechanical production of viable

protoplasts, although with low yields, is encouraging, and

supports the feasibility of developing an enzyme-mediated

method. Recently, crude extracts of the hepatopancreas of

Holiotus sp. have allowed preparation of protoplasts from

Sargassum hystrix (Preston et al., 1985b), although with

some inconsistency of yields which is not yet understood.

Fractionation of this type of crude extract and

identification of specific enzymatic components necessary

for protoplast formation may provide information which will

allow well defined bacterial and/or fungal enzymes to be

applied with more reproducible results.




Alginate lyases have been sought as specific probes to

study the structure of soluble alginates (Davidson et al.,

1976; Boyd and Turvey, 1978; Fujibayashi et al., 1970) and

polymers contributing to the structure of the cell walls of

the brown algae (Doubet and Quatrano, 1984). These enzymes

have recently been evaluated for the production of

protoplasts from Sargassum species (Preston et al., 1985b;

Romeo et al., 1986). Information about the substrate

specificities and the modes of substrate cleavage by

alginate lyases would be valuable in defining their

catalytic properties and their potential uses. Previous

studies on the mechanisms of alginate depolymerization have

included viscometric analyses (Kashiwabara et al., 1969;

Nakada and Sweeny, 1967; Elyakova and Favorov, 1974) and the

characterization of enzyme generated products utilizing a

number of methods including gel filtration (Boyd and Turvey,

1977; Favorov et al., 1979), ion exchange column

chromatography (Preiss and Ashwell, 1962a; Linker and Evans,

1984), paper chromatography (Preiss and Ashwell, 1962a;



Kashiwabara et al., 1969; Linker and Evans, 1984; Davidson

et al., 1977) and paper (Davidson et al., 1976; Davidson et

al., 1977), and gel electrophoresis (Hansen et al., 1984;

Doubet and Quatrano, 1984).

The isolation and characterization of alginate lyase

producing oxidative and fermentative marine bacteria

associated with actively growing tissues of marine brown

algae, genus Sargassum, have been described previously

(Preston et al., 1985a; Romeo et al., 1986). One of the

facultative anaerobes, isolate A (SFFB080483 A, see Table 2-

1), was shown to produce extracellular lyase activity which

was specific for poly(ManA) versus poly(GulA) and endolytic

in its action on alginate. The intracellular activities of

this bacterium depolymerized both poly(ManA) and poly(GulA)

and in comparison with the extracellular preparation showed

a greater ratio of bond cleavage to increase in fluidity

with native alginate, suggestive of exolytic as well as

endolytic mechanisms. Here the extracellular activity is

shown to belong to a single enzyme. A method based upon

HPLC separation of small oligomers is described, and it has

allowed a kinetic evaluation of the depolymerization of

poly(ManA) by both of these preparations. The limit

products generated by the poly(ManA) specific extracellular

enzyme have also been established.




Chemicals were analytical grade except as indicated.

Acetonitrile (Fisher Scientific) and tetrabutylammonium

hydroxide (Fisher Scientific and Sigma Chemical Co.) were

HPLC grade. Commercially available electrophoresis grade

reagents were used for electrophoretic analyses. All

aqueous solutions were prepared with water which was

deionized and glass distilled.

Preparation of Substrates

Sodium alginate was purchased from Fisher Scientific

Company as a purified grade originally isolated from

Macrocystis. Poly(GulA) and poly(ManA) were obtained from

HC1 hydrolyzed alginate following the methods developed by

Haug et al. (1967). These preparations were further

fractionated on Sephadex G-50 with 0.5 M NaCl as eluant, and

selected fractions were analyzed for reducing termini

(Nelson, 1944) and total carbohydrate (Dubois et al., 1956;

Haug and Larsen, 1962) to obtain substrates of uniform size.

Both 'H and 13C NMR (Grasdalen et al., 1979; Grasdalen et

al., 1981) analyses were carried out to assess the purity of

substrates. Poly(ManA) preparations contained 11%

guluronate, and poly(GulA) contained 11% mannuronate.


Enzyme Assays

Alginate lyase activity was quantified by spectrophoto-

metric determination at 548 nm of the chromophore formed

upon reaction of thiobarbituric acid (TBA) with periodate

treated products (Preiss and Ashwell, 1962; Weissbach and

Hurwitz, 1959). This method allows the specific measure of

unsaturated nonreducing termini of oligomeric products and

the unsaturated monomer, 4-deoxy-L-erythro-5-hexoseulose

uronic acid. Substrate mixtures contained either 0.1%

alginate, poly(GulA), or poly(ManA), and 0.05 M KC1,

buffered with 0.03 M sodium phosphate at the desired pH.

Enzyme was mixed with 9 volumes of substrate to start the

reactions, and reactions were terminated after 10 min by

addition of periodic acid solution. One enzyme unit is

defined as that amount of activity which will catalyze the

formation of 1 nmole of nonreducing termini and/or monomer

at pH 7.5 in one min at 220C. Protein was routinely

estimated by absorbance at 280 nm. For more quantitative

measurements the assay of Bradford (1976) was utilized with

bovine serum albumin as a standard protein.

Enzyme Isolation

The bacterium used in this study was obtained from

healthy, apical tissue of Sargassum fluitans and initially

identified as an organism which secretes alginate degrading

activity, based on the appearance of extensive clearing zones


surrounding colonies grown on solid alginate medium. The

organism used for the work described here has been desig-

nated isolate A (complete designation SFFB080483 A) and has

been described (Preston et al., 1985a; Romeo et al., 1986).

Biochemical and morphological properties of this bacterium

suggest its assignment to the genus Photobacterium, although

its DNA has a GC fraction of 0.454, somewhat greater than

that of other species currently included in this genus

(0.398-0.429). The organism has been maintained by monthly

transfer on solid alginate medium (Preston et al., 1985a).

For enzyme isolations the organism was grown in 0.1%

liquid alginate medium (Preston et al., 1985a) with rapid

gyrotory shaking at room temperature. In all subsequent

purification steps the extracellular and intracellular

preparations were kept at approximately 40C. Bacterial

cells were harvested at late exponential phase by

centrifugation at 10,000 x g for 10 min, washed twice with

water by resuspension and centrifugation, frozen in liquid

nitrogen, and stored at -700C. The spent medium was

concentrated and dialyzed against distilled, deionized water

by tangential flow filtration using a Millipore Pellicon

cassette system with a polysulfone membrane (PTGC) which

allowed retention of proteins larger than 10 kDa.

For extracellular enzyme preparations the alginate

lyase activity was precipitated along with remaining

alginate products by dropwise addition of 10% polytheyleni-

mine (PEI) to concentrated medium while stirring on ice.


The PEI was obtained as a 50% aqueous solution from Sigma

Chemical Co. Prior to use, this solution was diluted with

water, titrated to pH 7.5 with 12 N HC1, and centrifuged at

10,000 x g for 10 min to remove insoluble particles. The

relative volume of PEI necessary for maximal precipitation

of enzyme was found to be critical and was determined for

each batch of enzyme. Typically 1 ml of 10% PEI would yield

maximal precipitation of enzyme from 125 ml of concentrated

medium derived from 10 ul of spent medium. The precipitate

was collected by centrifugation at 10,000 x g for 15 min and

resuspended in distilled, deionized water using a Potter-

Elvehjem homogenizer driven by a variable speed motor. The

resulting suspension was centrifuged and the pellet

homogenized in 0.25 M NaC1, 0.1 M sodium phosphate at pH

7.5, to elute the enzyme. The suspension was centrifuged

for 1 h at. 150,000 x g, and the supernatant solution was

subjected to gel permeation chromatography on Sephacryl S-

200 (2.5 x 133 cm) with 0.1 M sodium phosphate at pH 7.0.

To obtain protein concentration sufficient for preparative

digestions and for quantification with the Bradford assay,

fractions were concentrated by ultrafiltration using an

Amicon cell with a YM 10 filter (10 kDa cutoff).

For intracellular enzyme preparations, cells were

thawed, suspended in 4 volumes of ice cold 0.1 M sodium

phosphate, pH 7.5, and disrupted with a French pressure cell

at 16,000 PSI. Unbroken cells and debris were removed by

centrifugation at 10,000 x g for 15 min, and the supernatant


solution brought to 2% streptomycin sulfate with the

addition of a 5% streptomycin sulfate stock solution. After

mixing for 10 min, the precipitate containing anionic

polymers was removed by centrifugation at 10,000 x g for 15

min and protein precipitated from the supernatant solution

with the slow addition of solid ammonium sulfate (to 65%

saturation). After centrifugation at 10,000 x g for 15 min,

the protein pellet, containing alginate lyase, was dissolved

in 0.1 M sodium phosphate buffer, pH 7.0, and chromato-

graphically fractionated on the Sephacryl S-200 column

(2.5 x 133 cm) with the same buffer.


The method for native gel electrophoresis was

described by Shuster (1971) and used the discontinuous

buffer system of Davis (1964). The running gel had a final

concentration of 7.5% acrylamide and 0.2% bisacrylamide, was

buffered at pH 8.9 with 0.38 M Tris HC1, and was polymerized

with 0.07% ammonium persulfate and 0.058% N,N,N',N'-

tetramethylethylene-diamine (TEMED). The stacking gel was

composed of 2.5% acrylamide, 0.5% bisacrylamide, 0.062 M pH

6.8 Tris HC1, and was polymerized with 0.058% TEMED and

0.01 mM riboflavin phosphate, using a fluorescent light to

activate the polymerization process. The running buffer was

composed of 0.3% Tris base, 1.44% glycine at pH 8.9.

Vertical slab gels were 0.15 cm in thickness and were

subjected to electrophoresis at 30 mA per gel until the


bromophenol tracking dye had reached the end of the gel.

The gels were cut into 0.5 cm slices which were incubated at

room temperature with 200 ul of sodium alginate pH 7.5

substrate mixture. Following the incubations 100 uL of the

solutions were withdrawn and assayed for products of the

lyase reaction.

Analysis of Alginate Lyase Generated Products

Analytical chromatographic separation of products was

accomplished by ion-paired reversed phase HPLC using a

system which is a modification of that developed by Voragen

et al. (1982) to fractionate pectate products. The column

was a C18 uBondapak 8MB 10 u column housed on a Z-Module

radial compression system (Waters). A Rainin 0.5 u

stainless steel filter and a Waters RCSS Guard-Pak C18

prefilter cartridge were positioned between the column and

the injector. The column was run at room temperature in an

isocratic mode with a solvent system of 10% acetonitrile,

10 mM tetrabutylammonium hydroxide, 0.1 M sodium phosphate

buffered at pH 6.5. Unsaturated oligomers were detected by

monitoring UV absorbance of the effluent from the column

at 230 nm with a Gilson Holochrome variable wavelength

detector equipped with a 1.0 cm flow cuvette. A Waters Tri-

Mod system was used for programming a 6000A pump,

integration of peak areas, and for automated injection of


samples onto the column. A Waters U6K injector was used for

manual injection of samples.

Preparative fractionation of products generated from

poly(ManA) by the extracellular enzyme was carried out by

gel filtration using a Biogel P-2 column (2.5 x 133 cm)

eluted with 0.1 M ammonium bicarbonate, collecting 5.8 ml

per tube. For a successful isolation of the products, 110

mg of poly(ManA) was dissolved in a 1.6 ml solution of the

extracellular enzyme (900 units/ml) buffered at pH 7.0 with

0.1 M sodium phosphate, and incubated for 12 h at room

temperature. Under these conditions the digestion was not

complete at the time the reaction mixture was applied to the

P-2 column, allowing some of the larger oligomeric

mannuronans to be obtained. Absorbance at 230 nm was

determined for each tube.

Contents of tubes comprising each peak from the P-2

column were pooled, lyophilized, and stored at -200C over

anhydrous calcium sulfate. The lyophilized products were

dissolved in 0.1 M sodium phosphate buffer, pH 7.0, and were

analyzed for unsaturated nonreducing termini by measuring

TBA reactive material generated by periodate oxidation

(Preiss and Ashwell, 1962a; Weissbach and Hurwitz, 1959) with

3-deoxy-D-manno-octulosonic acid (KDO, Sigma Chemical Co.)

as a standard. For preparation of the KDO standard, the

compound was desiccated overnight in vacuo. The chromophore

generated by the reaction of TBA with the B-formylpyruvate


formed by periodic acid oxidation of the KDO was quantified

spectrophotometrically according to Preiss and Ashwell

(1962) and according to Koseki et al. (1978) which indicated

that the sample was 90% and 85% pure, respectively. Total

uronic acid content was measured by the method of

Blumenkrantz and Asboe-Hansen (1973) using D-mannurono-

lactone (Sigma Chemical Co.) as a standard. Based upon the

expected extinction at 520 nm for the chromophore from

D-mannuronolactone, the desiccated standard was 84% pure.

Absorbance at 232 nm was measured in a 1.00 cm cuvette after

diluting samples 200-fold with 0.01 N HC1.

Samples of lyophilized fractions from the P-2 column

effluent were sent to Triangle Laboratories, Inc., Research

Triangle Park, North Carolina, for fast atom bombardment

(FAB) mass spectrometry under the direction of Ronald Hass.

Analyses were performed on a VG 7070H mass spectrometer with

a VGll-250 data system. The acceleration voltage was 3 kV

for the trimer analysis and 2 kV for the other samples. An

lontech saddle field ion source was used with xenon as the

bombarding species. The gun was operated at 7 keV with a

discharge current of ca. 1.5 mA. The samples were analyzed

after dissolving in water and applying 1-2 ul of the

solution to thioglycerol on the probe. The mass spectrometer

was scanned at 5 s per dec of mass from 1200-100, at a mass

resolution of 1000.


Results and Discussion

Chromatographic and Electrophoretic Behavior of the
Extracellular and Intracellular Activities

The extracellular preparation from a fermentative

marine bacterium, designated isolate A, was shown previously

to be highly active on poly(ManA) and native alginate, but

inactive or possessing only trace activity with poly(GulA)

(Preston et al., 1985a; Romeo et al., 1986). The

preparation was endolytic with alginate as substrate, as

shown by comparing the rate of bond cleavage with the

increase in the reciprocal of specific viscosity, i.e.,

specific fluidity. For purification and characterization of

extracellular poly(ManA) lyases, the concentrated

preparation was first treated with 10% PEI to remove the

partially degraded alginate which remained after dialysis of

the medium. This procedure is effective in reducing the

viscosity, allowing a greater quantity of enzyme to be

applied to the gel filtration column. When subjected to gel

filtration on Sephacryl S-200, a single peak of lyase

activity eluted at 0.67 column bed volumes (Fig. 3-la). The

column allowed complete removal of remaining products

derived from alginate, which appeared as TBA reactive

material eluting at 0.94 column bed volumes and separation

of some of the contaminating proteins from the enzyme. The

pooled enzyme fractions represented recovery of 82% of the

alginate degrading activity loaded onto the column. At this

stage of purification, the specific activity of the pooled


J 0.10



E 2.0

Figure 3-1.


L 0.000




40 60 80
Tube Number

Chromatographic fractionation of extracellular
and intracellular alginate lyase activities.
Samples of the extracellular and intracellular
fractions were obtained as described in the
text and subjected to chromatography at 4 C on
Sephacryl S-200. The column was eluted with
0.1 M sodium phosphate at pH 7.0, and 7.8 ml
fractions collected and assayed for protein by
absorbance at 280 nm, and for alginate lyase
activity by the TBA assay, absorbance at 548 nm.
The column was calibrated with molecular weight
standards, bovine serum albumin (BSA, 67 kDa),
S-lactoglobulin (B-L, 37 kDa), blue dextran
(to establish the void volume, Vo) and NaCl
(total column volume, Vt). The extracellular
preparation (a) was derived from medium which
yielded 3.7 g wet weight of cells and the
intracellular (b) from 13.8 g wet weight of


enzyme fraction was typically 3500 units per mg of protein

with alginate as the substrate and the ratio of activities

on poly(GulA) versus poly(ManA) was 0.14.

Samples from individual tubes containing the

extracellular alginate lyase fraction which eluted from the

Sephacryl S-200 column were subjected to native polyacryla-

mide gel electrophoresis followed by detection of activity.

The individual fractions comprising the alginate lyase peak

eluting from the S-200 column showed the same single

activity component which migrated as a homogeneous band

(Fig. 3-2a), indicating that the extracellular fraction

contained a single enzymatic activity.

Intracellular activities eluted from the same Sephacryl

200 column as two small and one large peak (Fig. 3-1b). The

major peak eluted in a volume corresponding to an estimated

molecular mass of 40 KDa and contained approximately 15% of

the alginate degrading activity originally loaded onto the

column. The enzymes present in the tubes from centers of

the three activity peaks were active on both poly(ManA) and

poly(GulA) and therefore were probably mixtures of two or

more alginate lyases. The ratios of activities on

poly(ManA) versus poly(GulA) were 6.8, 1.8, and 1.2 for the

contents of tubes 31, 37, and 44, respectively.

The peak tube from the major S-200 intracellular lyase

fraction was electrophoretically resolved into at least

three activities (Fig. 3-2b), indicating that the major

fraction eluting from this column included more than one


0 10 20
Slice Number

Figure 3-2.

Electrophoretic analysis of (a) extracellular
and (b) intracellular alginate lyase activities.
The extracellular preparation was fractionated
by gel filtration on Sephacryl S-200. A sample
was removed from the peak tube and subjected to
native polyacrylamide gel electrophoresis and
subsequent detection of activity according to
methods described in the text. A sample from
the peak tube from the Sephacryl S-200 frac-
tionated intracellular extract was subjected
to electrophoresis on the same native poly-
acrylamide slab gel used for analysis of the
extracellular lyase preparation.




( 0.00

C 0.30



alginate degrading enzyme. The activity with the highest

electrophoretic mobility could be identical to the single

major extracellular enzyme based upon electrophoretic

migration; however, this has not been confirmed by other


HPLC Analysis of Poly(ManA) Depolymerization

When the extracellular enzyme (14 units) from the S-200

column was incubated at room temperature with 5 mg of

poly(ManA) in 0.5 ml of 0.1 M sodium phosphate buffered at

pH 7.0, unsaturated oligomers were produced which could be

fractionated by HPLC. Profiles generated after 15 min and

4 h of depolymerization are shown in Figs. 3-3a and 3-3b,

respectively. At least six peaks were detected by

absorbance at 230 nm, and five of these were sufficiently

distinct to be integrated by the data analyzing system.

Individual oligomers, detected as A230 peaks (Fig. 3-3b),

were designated numerically from 1 to 5 in the order of

their elution with retention times (in minutes) of 5.87,

7.54, 10.00, 13.87, and 19.83, respectively. As will

be demonstrated, these represent the unsaturated dimer

through hexamer, respectively. When the depolymerization

was monitored over a 30 h period (Fig. 3-4) several features

of interest were noted. Component 1, the dimer, and to a

lesser extent component 2, the trimer, exhibited initial

lags in their rates of accumulation. Components 4 and 5,

the pentamer and hexamer, increased until the digestion had


0.04- a


0.00- --




3 10 20 3
Retention Time (min)

Figure 3-3.

Liquid chromatographic analysis of products
generated by digestion of poly(ManA) by the
extracellular lyase. The unsaturated oligomers
produced from poly(ManA) by the extracellular
enzyme were resolved by HPLC, as described in
the text. Sample volumes of 10 ul were
delivered with automatic injection (Waters,
WISP) and eluted isocratically at 1.0 ml/min.
Profiles of the products which had accumulated
after 15 min (a) and 4 h (b) are shown.



0 4.0-



Figure 3-4.


.o.-- ----- -o 3

.o-o-o- -o-
,,00 -

10 20 30
Time (h)

Kinetic analysis of poly(ManA) depolymeriza-
tion by extracellular alginate lyase. The
depolymerization reaction described in Fig.
3-3 was sampled periodically over 30 h, and
samples subjected to HPLC as described in the
text. The peak areas of the major products
are plotted against the times at which the
reaction was sampled. The individual products
were given number designations according to
their order of elution from the column, start-
ing with the fastest moving compound, 1,
representing a dimer, through 5, which repre-
sents a hexamer. The dimer peak integrates
as 0.11 area units per nmole.


continued for 6 and 4 h, respectively, and thereafter

decreased. The dimer, trimer, and tetramer never showed

declines, although the rates of accumulation of trimer and

tetramer decreased at approximately 6 h, and the rate of

dimer accumulation began to decrease gradually at 6 to 10 h.

The delays in appearance of the two smaller products, dimer

and trimer, indicated that these compounds were generated to

a significant extent from products which accrued from

initial depolymerization reactions. The decrease in the

concentration of the larger compounds, pentamer and hexamer,

after initial accumulation suggested that these must be

subject to depolymerization by the enzyme, and that their

relative rates of formation and degradation determine the

levels at any given time. The unsaturated monomer is not

readily detected by absorbance at 230 nm due to

tautomerization to the a-keto acid form (Preiss and Ashwell,

1962). Detection at 205 nm revealed a minor component which

eluted prior to the dimer, and analysis by the TBA method of

fractions collected from a reversed phase separation of a

preparative poly(ManA) digest also showed a minor component

of reactive material eluting prior to the dimer (data not

shown). This, presumed to be the monomer, represented less

than 4% of the products including dimers to pentamers

produced by the depolymerization of poly(ManA) catalyzed by

this enzyme.

The depolymerization of poly(ManA) by the intracellular

preparation was analyzed by HPLC using the methods described

for the extracellular preparation. Figure 3-5 shows the


products which have accumulated at 3 h and 21.5 h. During

an initial phase of the reaction, up to 8-10 h, four

products predominate. The mobilities of three of these are

identical to those of major products generated by the

extracellular enzyme, as established by direct comparisons

at the time of analysis (data not shown). Direct

comparisons were necessary due to changes in column

performance which occurred with usage. Products included

the trimer, with a retention time of 6.9 min, the tetramer,

with a retention time of 8.8 min, and the pentamer, with a

retention time of 11.5 min. The peak which was generated

only by the intracellular preparation presumably represents

the monomer compound, 4-deoxy-L-erythro-5-hexoseulose uronic

acid, and has a retention time of 4.7 min. The identity of

the monomer is indicated by its ratio of absorbance at 230

nm/205 nm, 0.16, as compared with those of the dimer through

tetramer, 0.9-1.3, and by its elution position, which is

prior to the dimer. Due to the poor absorption of the

monomer at 230 nm, the peak area for the monomer

underestimates its concentration relative to other products

several-fold. A fifth compound accumulates to a

considerable extent in the reaction after a lag of 10 h and

is seen in Fig. 3-5 in the lower profile. This product has

the retention time of the dimer compound, 5.95 min. Figure

3-6 shows that the dimer and trimer, peaks 2 and 3, continue

to accumulate at later times in the digest, long after the

concentrations of the monomer, peak 1, and tetramer, peak 4,

have become constant.


0.02 -


z In

M 0.04 -


0 10 20 30
Figure 3-5. Liquid chromatographic analysis of products
generated by digestion of poly(ManA) by the
intracellular alginate lyase activities. The
unsaturated oligomeric products generated from
poly(ManA), 10 mg/ml, by the third peak of
activity from the S-200 column, tubes 42-47,
13 units/ml, were resolved by HPLC as described
in the Materials and Methods. Profiles of the
products which had accumulated at 3 h, upper
profile, and 21.5 h, lower profile, are



o 2

S2.0 /

0 20 30
< ^r /
W. /
w //


0 10 20 30
TIME (h)

Figure 3-6. Kinetic analysis of poly(ManA) depolymeriza-
tion by intracellular alginate lyase activities.
The depolymerization reaction described in
Fig. 3-5 was monitored over 33.5 h, by
periodically subjecting 10 ul samples to
HPLC analysis. Products are numbered accord-
ing to their elution positions. Peak 1 is the
presumptive monomer, and peaks 2-4 are the
unsaturated dimer through tetramer.


The poly(ManA) lyase activity from the intracellular

preparation differs from the extracellular activity in

generating a considerable quantity of apparent monomer. The

physiological necessity to produce monomer on the inside of

a bacterial cell or at the cell surface is obvious, although

the mechanisms of the enzymes responsible for monomer

production in the reaction are not yet established. A

single exolytic enzyme might generate monomer as it degrades

poly(ManA) from either the reducing or nonreducing termini.

Alternatively, the monomer may be generated from some

intermediate degradation products by one or more enzymes

with glycosidase-like activities.

The unique appearance of the dimer product after a lag

of approximately 10 h is quite unexpected. The fact that

the preparation may contain more than one alginate degrading

activity does not allow a definitive explanation for this.

However, a number of possibilities might be envisaged. If

the monomer is being produced primarily from a dimer product

by a glycosidase-like activity, the loss of this activity

would be expected to allow dimer to accumulate. A second

possibility is that the dimer begins to accumulate as some

intermediate depolymerization product reaches a

concentration sufficient to allow its recognition by the

enzyme which is capable of generating dimer.


Purification and Characterization of the
Reaction Products

In order to obtain sufficient quantities of the

oligomeric products for characterization, 110 mg of

poly(ManA) was digested with extracellular enzyme for 12 h

and the products were resolved by chromatography on Biogel

P-2 eluted with 0.1 M ammonium bicarbonate. Four components

measured by absorbance at 230 nm were resolved from one

another. These were eluted at column volumes of 0.46, 0.50,

0.56, and 0.64 and designated as fractions 1, 2, 3, and 4,

respectively (Fig. 3-7). A small fraction, approximately 9%

of the total absorbance at 230 nm, eluted between the void

volume of the column and the leading edge of the peak

designated as fraction 1. Chromatography in ammonium

format led to a similar profile; however, the lyophilization

of these fractions resulted in discoloration of some of the

fractions and subsequent HPLC analysis demonstrated

significant degradation. Products obtained after elution

with ammonium bicarbonate were, after lyophilization, fluffy

and white, although they were quite hygroscopic.

Samples of the peak tubes from the P-2 ammonium

bicarbonate column were analyzed by HPLC (data not shown),

which identified the contents of tubes 52, 57, 63, and 72 as

the oligomers comprising peaks 4, 3, 2, and 1, respectively,

of the HPLC profiles (Fig. 3-3). Absorbance spectra in the

UV range for samples diluted in 0.01 N HC1 showed maxima



40 50 60 70 80

Tube Number

Figure 3-7. Preparative fractionation of the unsaturated
mannuronides. A 110 mg sample of poly(ManA)
was digested by 1,400 units of extracellular
enzyme in a 1.6 ml volume for 12 h at room
temperature and the products fractionated by
gel filtration with a P-2 column, as described
in the text. Volumes of 5.8 ml were collected
in tubes and assayed for absorbance at 230 nm.
Native alginate and galacturonic acid were
subjected to chromatography on the P-2 column
and their elution positions are indicated.
The contents of tubes comprising 4 fractions
were pooled and lyophilized: fraction 1,
tubes 52, 53; fraction 2, tubes 55-57; frac-
tion 3, tubes 62-64; fraction 4, tubes 71-73.



CO -



at approximately 232 nm (profiles not shown), typical of

products generated by alginate lyases.

The pooled, lyophilized fractions from the P-2 column

were further analyzed to establish the molecular sizes of

the unsaturated products. Comparisons of the uronic acid

content with content of nonreducing residues (Table 3-1)

allowed an estimation of the degree of polymerization (DP)

of each product. The DP values estimated using the ratio of

total uronic acid to nonreducing termini show a trend

consistent with assigning the unsaturated oligomers in

fractions 1 through 4 as the pentamer, tetramer, trimer, and

dimer compounds, respectively.

Based upon the concentration of the nonreducing

unsaturated terminal residues estimated with the TBA assay

of the periodate treated products and the abosrbance values

at 232 nm of products in 0.01 N HC1, an estimation of the

molar absorptivities for the dimer to pentamer series,

presumed to contain a single unsaturated residue in each

molecule, ranged from 5,160 (Fr 1) to 5,420 (Fr 3) M-cm-.

The concentrations of each component, based upon gravimetric

preparations of lyophilized samples and calculated molecular

weights of each as an ammonium salt, were calculated on the

assumption that fractions 1, 2, 3, and 4 represented the

pentamer, tetramer, trimer, and dimer, respectively. These

values, as divisors for the A232 values listed in Table 3-1,

led to calculated molar absorptivities which ranged from


Table 3-1.

Analysis of unsaturated oligomers from
lyophilized P-2 fractions.a

Fraction Abs.b Unsat.c Uronicd DP
no. 232 nm termini acids uronic
mM mM acids

1 39.2 7.59 40.5 5.3

2 42.0 7.94 37.5 4.7

3 64.0 11.8 36.3 3.1

4 84.6 16.2 45.6 2.8

aLyophilized products from the Biogel P-2 fractionated
preparative digest were obtained according to methods
described in the text and dissolved in 0.1 M sodium
phosphate buffer, pH 7.0, to a final concentration by
weight of 10 mg/ml.

bSolutions were analyzed for absorbance at 232 nm in a 1.00
cm quartz cuvette, after diluting 200-fold with 0.01 N HC1.

CUnsaturated nonreducing terminal residues were determined
by the TBA assay of periodate treated products, with KDO as
a standard. Values were adjusted to correct for the
apparent 90% purity of the standard.

dUronic acid residues were determined using D-mannurono-
lactone as a standard. Values were adjusted to correct for
the apparent 84% purity of our standard compound, based on
the expected yield of chromophore given by Blumenkrantz and
Asboe-Hansen (1973).


3350 (Fr 2 as the tetramer) to 3890 (Fr 1 as the pentamer).

The individual fractions, in particular the putative dimer,

were sufficiently unstable to heating to preclude high

temperature desiccation, and these lower molar absorptivi-

ties, in comparison to those determined from the concentra-

tions determined with the TBA assay, may reflect the

presence of water not removed by the lyophilization process.

Samples of unsaturated dimeric and tetrameric products

obtained from digestion of bacterial alginate with a

poly(ManA) lyase from a Pseudomonas aeruginosa isolate

(Linker and Evans, 1984) were graciously provided by Dr.

Alfred Linker and shown to possess the same HPLC mobilities

as our dimer and tetramer, respectively; molar absorptivi-

ties at 232 nm in 0.01 N HC1 of 6,400 and 5,500 M-1cm-1 were

obtained for this dimer and tetramer, respectively (data not


Further evidence of the DP of the major products was

obtained by FAB mass spectrometry (Table 3-2) of the

lyophilized fractions. Spectra from each of the products

contained major ions which correspond to within 1 mass unit

of the calculated M+NH4 and the M+NH4+H20 ions. The trimer

product showed an additional ion which represents the M+NH4

+ thioglycerol, and the spectrum of the dimer showed two

major ions (373, 391) which could not be explained, based on

the expected structure of the dimer. The dimer sample was

the only one which was not white in color, and we assume


Table 3-2. Analysis of unsaturated oligomers by FAB mass

Observed ions

M+NH4 +

Product M+NH4 M+NH4+H20 thioglycerol Unidentified

Pentamer 898 916

Tetramer 722 740

Trimer 546 564 654

Dimer 370 388 373, 391

aSamples of lyophilized fractions from the P-2 column were
analyzed by FAB mass spectrometry according to methods
described in the text.


that these ions resulted from some decomposition of the

dimer, which occurred in transit or in handling prior to FAB


Activity of the Extracellular Enzyme on Unsaturated
Oligomeric Products

Although the HPLC kinetic analysis of poly(ManA)

digestion yields valuable information about the reaction,

the complex nature of the process, wherein several products

compete for binding to the enzyme and some products are

degraded as they accumulate, does not allow detailed

consideration of the activity of the enzyme on individual

molecular species. To test the capability of the

extracellular enzyme to further degrade products which

accumulate during depolymerization of poly(ManA), the

lyophilized trimer, tetramer, and pentamer purified by P-2

column chromatography were individually incubated with

extracellular enzyme and the reactions sampled at 5 min and

5 h and subjected to HPLC. The conditions and the resulting

profiles are shown in Fig. 3-8. Profiles generated from

poly(ManA) digestion (al and a2) are included for

comparison, as the retention times for the products had

changed over several months of column use since the profiles

shown in Fig. 3-3 were obtained. It is clear that the

trimer is not subject to depolymerization by the enzyme, as

profiles bl and b2 are identical. The tetramer compound (cl

and c2) is not a good substrate for the enzyme, as predicted





CL1 c2 dl d2

0 0.04

field trimer, tetramer, and pentamer products.
The lyophilized products obtained after P-2
column chromatography were dissolved in 0.1 M
sodium phosphate at pH 7.0 such that the final
concentration in the reaction mixtures was
5 mg per ml, and enzyme was added to a final
concentration of 100 units per ml to start the
reactions. The mixtures were incubated at room
temperature and 5 ul samples withdrawn at
5 min and 5 h, and subjected to reversed phase
HPLC. The chromatographic profiles generated
from poly(ManA), trimer, tetramer, and pentamer
are designated a, b, c, and d, respectively,
and numbers 1 and 2 indicate sampling times
of 5 min and 5 h.
of 5 min and 5 h.


from the kinetic analysis (Fig. 3-4); however, a small

amount of trimer was generated from the tetramer over the 5

h period. An equal amount of monomer should also have been

produced, although it would not be readily detected by

absorbance at 230 nm as noted above. The pentamer was

readily degraded by the enzyme, which converted almost 50%

of the initial quantity to equal amounts of dimer and

trimer, but produced little or no tetramer in 5 h of


Few other bacterial poly(ManA) lyases have been

characterized to an extent which would allow a detailed

comparison with the extracellular enzyme analyzed in this

study. Doubet and Quatrano demonstrated that a cell-bound

enzyme from a marine bacterium could degrade poly(ManA) by

an apparent exolytic mechanism (Doubet and Quatrano, 1984).

Davidson et al. (1977) described an endolytic poly(ManA)

lyase which was induced by phage infection of Azotobacter

vinelandii. This enzyme seems to be quite similar to the

one which we have studied, in that it generates a series of

unsaturated products ranging from dimers through pentamers,

although neither the relative levels of the products nor the

limit products were determined. Kashiwabara et al. (1969)

measured poly(ManA) (SM) degrading activities in crude

extracts of two marine pseudomonads. The activities were

weak in relation to the endogenous poly(GulA) lyase

activities, and although the reaction products were not well

characterized, an unsaturated trimer was shown to be the


major product. Linker and Evans (1984) examined an

intracellular poly(ManA) lyase from a Pseudomonas aeruginosa

isolate which generated unsaturated oligomers ranging from

dimeric through pentameric compounds. This enzyme was

apparently incapable of producing monomer and cleaved the

unsaturated tetramer to form dimeric products. Although the

major reaction products generated by the enzyme which we

have studied are similar to those produced by the P.

aeruginosa enzyme, the catalytic mechanisms of the two

enzymes clearly differ, as shown by the HPLC analysis of the

conversion of unsaturated tetramer to trimer.

The approach utilizing reversed phase ion-pairing HPLC

to evaluate the mechanisms of the lytic depolymerization of

alginate is being applied to study alginate lyases with

different substrate specificies, such as the extracellular

preparation from isolate G, and enzymes from other marine

bacteria (work in progress, J.F. Preston and T. Romeo).

This method should also prove useful for the study of lyases

acting on other glycuronans. Detection of products based on

refractive index or absorbance of UV at shorter wave

lengths (e.g., 205 nm) should extend the applicability of

the method to hydrolytic systems as well.




Previous studies utilizing alginate lyases have

examined the structure of alginate (Min et al., 1977; Boyd

and Turvey, 1978), the composition of alginate containing

cell walls of brown algae (Quatrano and Peterman, 1980), and

the feasibility of generating protoplasts of brown algal

species (Preston et al., 1985b; Romeo et al., 1986). The

possibility that the alginate produced by Pseudomonas

aeruginosa strains colonizing the lungs of cystic fibrosis

patients is involved in the morbidity of that disease has

recently led to the identification of alginate lyases in

isolates of clinical origin (Linker and Evans, 1984; Dunne

and Buckmire, 1985).

With few exceptions alginate lyase enzymes have been

examined as impure mixtures of proteins, or even as

preparations containing more than one activity, disallowing

firm conclusions to be drawn about their substrate

specificities, mechanisms, and their structural properties.

The result is that the only investigations on the structures

of these enzymes, with the exception of molecular weight



determinations, have been carried out on two isozymes from

the mid-gut gland of the wreath shell, Turbo cornutus

(Muramatsu and Egawa, 1982; Muramatsu et al., 1984).

We previously reported the isolation of an

extracellular alginate lyase capable of depolymerizing

poly(1-4)--D-mannuronan, poly(ManA), derived from alginate,

and an analysis of the products of this enzymatic reaction

(Chapter III). A method for purification of the

extracellular enzyme to homogeneity using HPLC is now

described. Structural properties of the enzyme which have

been examined include the molecular mass, pi, amino acid

composition, content of helical secondary structure, and the

N-terminal amino acid sequence. Some of the properties are

compared with those of other alginate depolymerizing


Materials and Methods


Chemicals were analytical grade except as indicated.

Commercially available electrophoresis grade reagents were

used for SDS polyacrylamide gel electrophoresis. Reagents

and chemicals for amino acid analysis and N-terminal

sequencing were commercially available ultrapure grades.

Water for all aqueous solutions was deoinized and glass



Sodium alginate was purchased as a purified grade

(Fisher Scientific Co.) originally obtained from

Macrocystis. The content of mannuronic acid was determined

to be 67% by 1H NMR, using methods established by Penman and

Sanderson (1972) and Grasdalen et al. (1979). The

poly(ManA) was prepared from HC1 hydrolysed alginate

according to Haug et al. (1967) and fractionated by size

through Sephadex G-50 with 0.5 M NaCl as eluent. The

fraction used for these studies was shown to contain 89%

mannuronic acid by 1H NMR. Comparison of total carbohydrate

(Dubois et al., 1956) to reducing termini (Nelson, 1944)

indicated that the range for the degree of polymerization

was 16-22.

Enzyme Assays

The poly(ManA) lyase activity was quantified by

spectrophotometric determination of the chromophore

generated upon reaction of thiobarbituric acid, TBA, with

periodate treated products (Preiss and Ashwell, 1962a;

Weissbach and Hurwitz, 1959). The following conditions have

been used for the enzyme reactions, unless otherwise noted:

pH 7.5, 0.03 M sodium hydrogen phosphate, 0.05 M KCl, 0.10%

sodium alginate, incubated for 10 min at room temperature,

22 C. One unit of activity will generate one nmole of

unsaturated termini and/or unsaturated monomer in 1 minute.

The quantity of protein present in fractions at various

stages of purification was determined by the Coomassie blue


binding assay of Bradford (1976), using bovine serum albumin

as the standard. When a more accurate estimate of the

protein concentration of the purified poly(ManA) lyase was

needed, spectrophotometric analysis at 205 and 280 nm was

used (Scopes, 1974). This method, unlike that of Bradford

(see Tal et al., 1985; Compton and Jones, 1985), has

relatively little variation of response to proteins of

differing chemical constitution.

Purification of Poly(ManA) Lyase

The poly(ManA) lyase was purified from a bacterium

originally isolated from healthy tissues of Sargassum

fluitans; the bacterium grew on alginate as sole

carbon source and secreted significant alginate lyase

activity. The properties of this fermentative marine

bacterium, designated as isolate A, or SFFB080483 A, have

been described (Preston et al., 1985a; Romeo et al., 1986).

The purification steps were carried out at 4 C, except for

chromatography in the HPLC systems, which was at room

temperature. For enzyme isolations the organism was grown

to late exponential phase in 0.1% liquid alginate medium

(Preston et al., 1985a), with rapid gyrotory shaking at 22 C.

Bacterial cells were separated from the medium by

centrifugation (10,000 x g, 10 min), and the medium was

concentrated and dialyzed against water by tangential flow

filtration using a Millipore Pellicon Cassette System with a


polysulfone membrane (PTCG) which allowed retention of

proteins larger than 10 kDa.

The enzyme was precipitated along with remaining

products of alginate degradation by dropwise addition of 10%

polyethylenimine (PEI, purchased from Sigma Chemical Co.,

St. Louis, MO, titrated to pH 7.5 with concentrated HC1,

diluted with water, and centrifuged at 10,000 x g to remove

insoluble particles) to the concentrated medium. The

relative volume of PEI needed for maximal precipitation of

alginate lyase activity was determined by titrating soluble

alginate lyase activity. The PEI precipitate was collected

by centrifugation (10 min at 10,000 x g) and washed by

resuspension in water with a Potter-Elvehjem homogenizer

driven by a variable speed motor. The suspension was

centrifuged and enzyme activity eluted from the precipitate

by homogenization in 0.25 M NaC1, 0.1 M sodium hydrogen

phosphate at pH 7.5. Insoluble material was removed by

centrifugation at 150,000 x g for 2.5 h, and the supernatant

solution subjected to gel filtration chromatography.

A preparation which was derived from 43 1 of growth

medium (yielding 71.3 g of wet cells) was fractionated on a

column of Sephadex G-75 superfine grade (5 x 79 cm) at 4 C

with 0.1 M sodium phosphate buffered at pH 7.0. Fractions

were collected (7.8 ml) and assayed for enzyme activity and

protein absorbancee at 280 nm). The recovered activity was

concentrated by pressure filtration in an Amicon stirred

cell (Amicon Corp., Lexington, MA) with a YM 10 membrane.


The concentrated activity was applied in 3 separate

runs to a Mono Q HR 5/5 anion exchange column (5 x 50 mm,

Pharmacia, Inc., Piscataway, NJ) and eluted at room

temperature with a gradient of NaCl (0-1.0 M) buffered at pH

7.0 with 0.01 M sodium hydrogen phosphate at a flow rate of

0.5 ml/min. The chromatography system included an LKB

Ultrachrome GTi HPLC system (2152 Controller, 2150 Pump,

2154-002 Injector, LKB-Produkter AB, Bromma, Sweden). A

Gilson Holochrome variable wavelength detector fitted with a

1.00 cm cell (Gilson Medical Electronics, Inc., Middleton,

WI) was used to measure absorbance at 280 nm, which was

recorded with a Linear 800 Versagraph (Linear Instruments

Corp., Irvine, CA). Fractions of 0.5 ml were collected and

assayed for alginate lyase activity.

Enzymatically active fractions from the 3 Mono Q column

runs were concentrated and desalted using the Amicon cell

and again applied to the Mono Q column and eluted as above.

Activity was recovered and concentrated and subjected to gel

filtration HPLC using an UltroPac TSK-G4000 SW column (7.5 x

600 mm, LKB) run at room temperature with a buffer of 0.1 M

sodium hydrogen phosphate, pH 7.0, containing 0.1 M NaC1, at

a flowrate of 0.2 ml/min. The column was fitted to the HPLC

system described above.


The SDS polyacrylamide gel electrophoresis was

performed (Laemmli, 1970) using single dimension 1.5 mm


thick slab gels, 9.73% acrylamide, 0.27% bisacrylamide, pH

8.8 for the running gel and 3.85% acrylamide, 0.11%

bisacrylamide, pH 6.8 for the stacking gel. The conditions

for the analyses are described in detail in the Hoefer

Scientific Instruments Catalog (Hoefer Scientific

Instruments, San Francisco, CA).

Isoelectric focusing gels were purchased as 1.0 mm

thick prepared gels, Ampholine PAGplates pH 3.5-9.5 (LKB)

and were run using an LKB Multiphore system according to

instructions provided with the gels. The isoelectric point

of the enzyme was determined by comparison with standard

proteins. Enzyme activity in the gel was determined by

sectioning the gel into 0.5 cm slices which were incubated

overnight with 200 ul of alginate under standard conditions.

Samples (100 ul) of the solutions were withdrawn and assayed

for unsaturated products. Proteins present in both SDS and

isoelectric focusing gels were fixed with acetic

acid/ethanol/water (1:5:4) and were visualized by staining

at 65 C for 30 min with Coomassie brilliant blue R-250, 0.46

g/400 ml of destain solution (acetic acid/ethanol/water,

1:2.5:6.5) and destaining of the gels with several changes

of destain solution.

Circular Dichroism Spectroscopy

Analyses were performed using a Jasco J500C

Spectropolarimeter. The scan speed was 20 nm/min, at a

sensitivity of 1 mdeg/cm using a spectral bandwidth of 1 nm


and a time constant of 2 sec. The samples were contained in

a 0.1 cm pathlength cuvette. Data processing was

accomplished with an Oki IF 800 Model 30 computer to provide

scan averaging and molar elipticity values.

Amino Acid Composition

Purified alginate lyase was dialyzed against water and

concentrated using a Centricon-10 unit (Amicon), lyophilized,

and hydrolyzed in 6 N HC1 under N2 in sealed tubes for 24 h

at 100 C. The amino acids were resolved and quantified

using a Beckman 6300 Amino Acid Analyzer with a Nelson

Analytical Data Acquisition System. The amino acid analyses

were carried out by B. Parten and B. Dunn in the Department

of Biochemistry at the University of Florida.

The content of cystine plus cysteine was obtained after

hydrolysis in the presence of dimethylsulfoxide, which

converts these amino acids to cysteic acid (Spencer and

Wold, 1969). Tryptophan content was estimated after

hydrolysis in the presence of 4% thioglycolic acid

(Matsubara and Sasaki, 1969).

Serine and threonine are known to be degraded slightly

under the conditions of hydrolysis, and their levels were

estimated by extrapolation to 0 h of hydrolysis from their

levels at 24 and 48 h of hydrolysis (Hirs et al., 1954).


N-Terminal Amino Acid Sequence

Sequence analyses were carried out by B. Parten and B.

Dunn. An Applied Biosystem Model 470A Gas Phase Protein

Sequencer was used for automated Edman degradation. The

program (02RPTH) ran 30 cycles with 2 nmole protein. The

repetitive yield was 94% and the initial yield was 70-80%

with a myoglobin standard. The PTH-amino acids were

identified and quantified using reversed phase HPLC with a

Waters Model Trimod system including a 721 Programable

System Controller, 730 Data Module, and a WISP 710B

automatic injector, using a Waters Model 440 Absorbance

Detector to monitor absorbance at 254 nm. A Novapak C-18

column (3.9 x 150 mm) was eluted with a gradient of

methanol, 10-90% in 0.025% acetic acid, to resolve the PTH-

amino acids.


Purification of Poly(ManA) Lyase

Table 4-1 summarizes the purification procedures. The

first step in the purification of the enzyme from the crude

extract, precipitation with PEI (see Fig. 4-1 for titration

of soluble activity) followed by'elution of enzyme activity

with 0.1 M sodium hydrogen phosphate containing 0.25 M NaC1,

separates a large amount of acidic carbohydrate from the

enzyme activity, decreases the viscosity of the preparation

and allows a greater amount of activity to be fractionated

on the Sephadex column.





0 )0

H> E

(U w



0 W
(0 ct4- 3

E--0 U






o M r O o


I *

o C C0 of

o o 0 0
o 0 '.0 -4 aO
I m N CN
I -

S0 0 0 0 0
o o 0 0 0 0

4 0 0

3 ( I I
0) ) O (0
cd 0 0

3 H 04 C C M
M H 0) 0 0 1
U C U E S E-

4 Nm M -3 L


4 -


ro w



c -

c m






C -4

0 0
3 a


O m
w u

En 0

0 C

O w



0 E


a) 0)
4-1 .c
0 4-'
o ct


W --
4 0)
(0 4-J

c (0
(0 V
4-) >1

a) .C

(0 0

E u
3 10

I4 O

0 0*
fl o
> 4-
0) 0 C

> .c
0 C


M (04-1

w O

0 U

O 0V

0 C
4-i 0)0)
> O

C 4-M

0 410
0 (0

V (00)

(0 UV

X 0




U 0.25 -


Figure 4-1.

2.0 4.0 6.0
PEI (p.I1)

Titration of soluble alginate lyase activity
with addition of 10% polyethyleneimine. A
crude extracellular preparation was concen-
trated 300-fold and dialyzed against water
using the Pellicon system, as described in the
Materials and Methods. Samples (100 ul) of
the preparation were vortexed at 0 C with
added volumes of PEI, the precipitates were
removed by centrifugation (13,000 x g, 5 min),
and the supernatant solutions (10 ul samples)
were assayed for enzyme activity absorbancee
at 548 nm).


Chromatography on Sephadex G-75 separated alginate

lyase activity as a single peak at approximately 0.5 column

volumes (see Fig. 4-2) from some of the larger contaminating

proteins which eluted in the void volume of the column.

Remaining alginate degradation products eluted near the

total column volume (data not shown) and were detected by

absorbance at 232 nm and oxidation with periodate to TBA

reactive compounds. The enzyme activity was completely

dependent upon the addition of exogenous substrate after

this step.

Anion exchange HPLC using a Mono Q column afforded a

purification of 27-fold. Less than 1 h was required for

recovery of the enzyme. The conditions shown in Fig. 4-3

allowed optimum separation of enzyme activity, which eluted

at approximately 0.4 M NaC1, from proteins with similar

charge properties. The procedure was repeated one time with

slight (1.8-fold) improvement in specific activity (Fig.


A final step in the purification was gel filtration

HPLC using a Biosil TSK-G4000 SW column (Fig. 4-4, panel a).

Although this procedure yielded only 1.04-fold purification

(Table 4-1) of the enzyme for the preparation considered

here, in purification of other batches of enzyme the fold

purification was as high as 1.2-fold, and the step has

been used routinely in the purification sequence. Samples

containing up to 250 ug of protein from the Mono Q column

have been successfully purified with the TSK column, and the





Figure 4-2.



0 50 100 150

Fractionation of alginate lyase activity by
Sephadex G-75 chromatography. A preparation
of activity which was obtained from the PEI
procedure was subjected to gel filtration
chromatography using Sephadex G-75, as
described in the Materials and Methods.
Enzyme activity (A548) and a relative measure
of protein (A280) were determined for frac-
tions which were collected, and the fractions
containing alginate lyase activity were pooled
and concentrated by pressure ultrafiltration
with an Amicon stirred cell.






' 0.2



Figure 4-3.

16 32 48 64
Anion exchange HPLC of poly(ManA) lyase. A
preparation containing 730 units of activity
and 270 ug protein from the Sephadex G-75 column
was fractionated with a Mono Q column using a
gradient of NaCI (up to 1.0 M, top portion of
gradient not shown) to elute the activity (panel
a). The single peak which possessed alginate
lyase activity is indicated by an arrow.
Activity peaks from three column runs were
pooled, concentrated and desalted, and this
combined preparation was subjected again to
anion exchange chromatography with the Mono Q
column (panel b).



CV 0.01



Figure 4-4.

60 120 1 2 3

Gel filtration HPLC and SDS-polyacrylamide gel
electrophoresis of poly(ManA) lyase. A sample of
lyase activity (10 ug of protein) from the second
pass through the Mono Q column was analyzed by gel
filtration HPLC on a TSK4000 column as described
in the Materials and Methods section (panel a).
The elution of standard proteins shown for com-
parison are as follows: 1, 8-amylase, 200 kDa;
2, bovine serum albumin, 66 kDa; 3 egg albumin,
45 kDa and B-lactoglobulin, 37 kDa; 4, carbonic
anhydrase, 29 kDa; 5,trypsinogen, 24 kDa; 6,
lysozyme, 14.3 kDa.

Activity from the second pass through the
Mono Q column was denatured in the presence of
SDS and 2-mercaptoethanol and analyzed in the
presence and absence of standard proteins (panel
b). Lane 1 contains standard proteins only,
lane 2 contains 10 ug poly(ManA) lyase, and
lane 3 contains lyase plus standards.


method has allowed the detection of minor contaminants which

were not apparent in the SDS polyacrylamide gel analyses

(data not shown). The poly(ManA) lyase was judged to be

pure when enzyme activity eluted as a single homogenous peak

of UV absorbing material (280 nm) on gel filtration HPLC, a

single band was observed on SDS gel electrophoresis (see

Fig. 4-4 panel b), and a specific activity of approximately

48 units/ug (as calculated in Table 4-1) was obtained.

Molecular Mass of the Active Enzyme and Its
Single Subunit

The molecular mass of the native enzyme was estimated

by gel filtration HPLC (Fig. 4-4, panel a; Fig. 4-5, panel

a) to be 29 kDa. Polyacrylamide gel electrophoresis in the

presence of SDS showed a single band at approximately 29 kDa

(Fig. 4-4, panel b; Fig. 4-5, panel b) indicating that the

active enzyme is composed of a single polypeptide.

UV Spectrum

The absorbance spectrum of the purified enzyme shows a

maximum at 280 nm, a minimum at 250 nm, a ratio of

absorbance at 280 to 250 of 2.6 (Fig. 4-6), and no measurable

absorbance in the visible range (data not shown). The

absorbance at 280 nm of a 1 mg/ml solution of enzyme at pH 7

was 1.6 when the protein concentration is calculated by the

method of Scopes (1974). The dye binding assay of Bradford

(1976), using bovine serum albumin as a standard protein,